STAR - Plastic Concrete For Cut-Off Walls PDF

State-of-the-Art Report
June 2018
Plastic Concrete for Cut-Off Walls
by
Karlsruhe Institute of Technology (KIT)

Institute of Concrete Structures and Building Materials (IMB)
Materials Testing and Research Institute (MPA Karlsruhe)
David Alós-Shepherd, M.Sc.

Dr.-Ing. Engin Kotan
Univ.-Prof. Dr.-Ing. Frank Dehn
KIT – The Research University in the Helmholtz Association

Please cite the following document as:
Alós Shepherd, D.; Kotan, E.; Dehn, F. (2018):

State-of-the-Art-Report - Plastic Concrete for Cut-Off Walls,
Institute of Concrete Structures and Building Materials (IMB),
Materials Testing and Research Institute (MPA Karlsruhe),
Karlsruhe Institute of Technology. Karlsruhe.
DOI: 10.5445/IR/1000085901
Abstract
Plastic Concrete plays a key role in the remediation of earthen dams using cut-off
walls to counter dam seepage. However, Plastic Concrete has yet to be thor-
oughly understood, since little attention has been given to this material in litera-
ture. The principal objective of this report is to set out the fundamental material
science parameters, which describe Plastic Concrete’s mechanical and hydraulic
behaviour as well as describing the mix design and application of Plastic Con-
crete. For this, an extensive and comprehensive literature review was carried
out. The results show that Plastic Concrete can hereby be considered to be a
low-strength, low-stiffness impervious concrete with a high deformation capacity
under load and the capability of sustaining larger strains than normal concrete.
This study further identifies reference values, which may be used in cut-off wall
design. All in all, the research results represent a further step towards the under-
standing of Plastic Concrete material behaviour.
iii
Table of Contents
Abstract iii
1 Introduction 1
1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Definition & Field of Application . . . . . . . . . . . . . . . . . . . . 2
1.3 Cut-Off Wall Construction . . . . . . . . . . . . . . . . . . . . . . . 3
1.4 Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.5 Problem Defintion . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2 Mix Design 8
2.1 Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.1.1 Cement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.1.2 Bentonite . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.1.3 Aggregates and Admixtures . . . . . . . . . . . . . . . . . . 15
2.2 Mixture Composition . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.3 Mixing Sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.4 Fresh Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3 Mechanical Behaviour 21
3.1 Testing Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.1.1 Compressive Strength . . . . . . . . . . . . . . . . . . . . . 21
3.1.2 Elastic Modulus . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.1.3 Testing Influences . . . . . . . . . . . . . . . . . . . . . . . 25
3.2 Plastic Concrete Strength . . . . . . . . . . . . . . . . . . . . . . . 27
3.2.1 Unconfined Compressive Strength (UCS) . . . . . . . . . . 28
3.2.2 Tensile Strength . . . . . . . . . . . . . . . . . . . . . . . . 33
3.2.3 Multi-Axial Load-Bearing Capacity . . . . . . . . . . . . . . 35
3.3 Elastic Modulus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
3.4 Creep and Relaxation . . . . . . . . . . . . . . . . . . . . . . . . . 42
v
PLASTIC CONCRETE FOR CUT-OFF WALLS TABLE OF CONTENTS
4 Hydraulic Behaviour 46
4.1 Testing Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
4.1.1 Unloaded Conditions . . . . . . . . . . . . . . . . . . . . . . 46
4.1.2 Loaded Conditions . . . . . . . . . . . . . . . . . . . . . . . 47
4.2 Plastic Concrete Permeability . . . . . . . . . . . . . . . . . . . . . 49
5 Summary and Outlook 53

5.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
5.1.1 Mix Design . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
5.1.2 Mechanical Behaviour . . . . . . . . . . . . . . . . . . . . . 54
5.1.3 Hydraulic Behaviour . . . . . . . . . . . . . . . . . . . . . . 55
5.2 Future Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
Acknowledgements 58
Bibliography 59
vi
1. Introduction
1.1 Background
The worldwide aging infrastructure is a reason for concern in many countries.
Unfortunately, only when a catastrophic failure of some infrastructure occurs, this
topic obtains public awareness. A key example for the systematic, catastrophic
failure of embankment dams and levees occurred in 2005 during the Katrina and
Rita Hurricanes in the North-American Gulf Shore area [33]. Most recently, in
February 2017, the Oroville Dam Failure further emphasized the critical situa-
tion of many dam infrastructures [75]. These catstrohpes have highlighted the
need for remediation works on a worldwide scale. In the United States alone,
approximately 91,000 dams are currently in need of some type of repair in vary-
ing degrees of deterioration [33, p.1] [126]. Furthermore, approximately 86% of
these dams are earthen dams, where the average age of these dams is currently
50 years [126]. Various failure modes are possible for earthen dams, ranging
from dam over-topping and inadequate maintenance to foundation defects and
slope instability. The latter generally occurs through water seepage below the
dam body causing a reduction in internal friction and causing the dam to slip.
Moreover, seepage may cause piping within the dam and even a hydraulic heave
failure to occur. Therefore, major concern is raised regarding dam safety and
various dam repair and remediation programs have been initiated worldwide.
A common solution to counter dam seepage is the design and construction of
cut-off walls [74]. This may occur in the design phase of new earthen dams or be
constructed during remediation of an existing dam. Figure 1.1 shows a possible
design concept.
The planned cut-off wall is hereby extended into an underlying impervious
stratum, e.g. rock [125, p.46]. For the cut-off wall construction and material
choice various possibilities exist. Cut-off walls may be constructed with mixed-in-
place technologies, grouting methods or through excavated and backfilled cut-off
constructions [33]. Each of these construction methods has its advantages and
disadvantages, for which the appropriate choice of method greatly depends on,
1
PLASTIC CONCRETE FOR CUT-OFF WALLS INTRODUCTION
Figure 1.1: Schematic of an Earth Dam with Cut-Off Wall
amongst others, time, cost, geographical location, technological and geological

factors [26, 33]. The most effective cut-off walls for seepage control can be con-
structed with excavated slurry-trench walls, especially for greater depths [74,133].
For these, a wide range of backfill materials may be used depending on specific
project requirements. For this reason, excavated cut-off walls are of preferred
choice for modern dam remediation. As backfill materials a wide range of possi-
bilities exist, e.g. standard concrete, soil-cement, soil-cement-bentonite, cement-
bentonite or Plastic Concrete [85, p.VIII-2] [125, p.16-7].
1.2 Definition & Field of Application

Generally Plastic Concrete can be considered to be a low-strength, low-stiffness
impervious concrete with a high deformation capacity under load. The European
standard EN 1538 [47] defines Plastic Concrete as a low-strength, low Young’s
modulus concrete capable of sustaining larger strains than normal concrete. This
material should have high deformability and low permeability while ensuring suf-
ficient material workability and strength [47, p.19]. Similarly, the DWA M512-1
guideline [56] considers Plastic Concrete to be a material which has a low hy-
draulic conductivity and high deformability. This guideline also gives reference
values and recommendations for cut-off walls in general. The United States Bu-
reau of Reclamation’s Design Standard No. 13 [125] states that Plastic Concrete
is a regular concrete which, through the addition of bentonite, becomes less stiff
and can therefore undergo greater strains before cracking compared to usual
concrete walls [125, p.16-47].
Plastic Concrete is furthermore placed under a supporting fluid (e.g. ben-
tonite suspension) using the so-called tremie method. With this, Plastic Concrete
has less potential for construction defects than the conventional soil-bentonite or
2
cement-bentonite backfilling [72]. In addition, the high strain capacity of Plastic

Concrete is of great advantage when ductile walls are needed if unequal deforma-
tions of the cut-off wall, large annual reservoir fluctuations or significant seismic
events are expected, causing significant bending strains to be induced [33] [35,
p.570] [125, p.58]. Through the highly ductile behaviour of the material, the rup-
ture probability can be decreased and wide, open cracks can be reduced to a
minimum hindering a material permeability increase [33, p.230] [77]. In specific
situations Plastic Concrete can also be used for the containment of contaminated
soils and other applications [27, 130, 136]. Plastic Concrete has therefore been
widely used in dam remediation for many years, with projects like the Sylven-
stein Dam (Germany) [95], Hinze Dam (Australia) [24] or Bagatalle Dam (Mauri-
tius) [25].
1.3 Cut-Off Wall Construction

As mentioned previously, the main task of a cut-off wall is to mitigate seepage,
generally below an existing or new dam. With the ever evolving technology in
specialist foundation engineering depths in excess of 120 m and ground strengths
greater than 160 MPa can be safely excavated [31, p.305]. The most common
method is the excavation of a slurry-trench wall with the use of clamshells or
cable-suspended hydro-cutters, as shown in Figure 1.2, 1.
Figure 1.2: Construction of a Cut-Off Wall with the Two-Phase Method [27]
The excavating method performance and choice greatly depends on depth,

ground strength and boulder content [31, p.308] [124]. For example only the
cable-suspended hydro-cutters are capable of greater depths and can be main-
tained under tight tolerances [68] [125, p.46]. During excavation the panels are
continuously filled with a bentonite or polymer slurry to hold the trench in place.
3
This working slurry is then cleaned (de-sanded) or replaced by a fresh slurry

to ensure homogeneous conditions for tremie concreting [66] as shown in Fig-
ure 1.2, 2 . This slurry is then replaced with Plastic Concrete using the tremie
placement method, whereby concrete is fed through hoppers into metal pipes re-
placing the slurry from bottom to top [125, p.46], as shown in Figure 1.2, 3 . This
also clearly differentiates Plastic Concrete cut-off walls from cement-bentonite
walls since the latter are produced in a single phase with self-hardening slur-
ries without the use of the tremie method [55, 124]. Only through the use of the
two-phase method, panels with greater depth are even possible due to the long
excavation times and increases suspension of soil [86, p.N28].
Concrete placement using the tremie method must however be done with great
care since various influencing factors exist [70]. The bentonite cake may for ex-
ample remain in the construction joints and therefore adequate construction ex-
pertise is essential for the correct construction of cut-off walls [31, p.309]. The
excavated panels are generally between 2.50 m and 7.50 m in length [124, p.37].
The panels width ranges between 0.6 m for low depths and 1.2 m for greater
depths. The added width is generally required to assist in maintaining overlap
between adjacent panels and ensure imperviousness of the cut-off wall [74]. To
construct the complete cut-off wall, construction is planned in an alternating se-
quence of primary and secondary panels, as shown in Figure 1.2, 4.
Although the slurry-trench method is the most widespread excavating method,
for small depth cut-off walls Plastic Concrete has also been placed with the
secant pile method. For example, the 30 m deep cut-off wall at the Papadia
Dam (Greece) was constructed using this method, whereby a minimum pile over-
lap of 0.7 m was used to ensure cut-off wall imperviousness [3, p.1107].
1.4 Requirements
To ensure the imperviousness of Plastic Concrete cut-off walls, these walls must
meet various requirements. Excluding cost, anticipated hydraulic gradient across
the cut-off wall and depth of cut-off wall are two of the main consideration in-
fluences [125, p.8]. The mix design of cut-off wall materials is hardly standard-
ised, since regulatory authorities are less concerned with the nature of the ma-
terial, than with its performance [86, N13]. Plastic Concrete however, has many
more configuration possibilities than cement-bentonite, which is a distinct class
of material with a distinct range of typical properties [86, N13]. Most commonly,
standards and guidelines limit the compressive strength, Young’s modulus and
hydraulic conductivity of Plastic Concrete.
4
Compressive Strength Requirements
For the required Plastic Concrete compressive strength various standards and
guidelines give varying requirements. ICOLD Bulletin No. 51 recommends the
use of the lowest compressive strength possible, to obtain a material having the
highest possible deformability [87, p.11]. Similarly, the DWA M512-1 guideline
requires a minimal unconfined compressive strength of 0.3 MPa at 28 days to
account for sufficient erosion resistance [56]. The compressive strength require-
ments do not necessarily have to be at 28 days, since Plastic Concrete strength is
known to increase strongly over time (as will be described in subsection 3.2.1) and
is not required for construction purposes. For this reason DIN EN 1538 suggests,
that long term strength and deformability may be accounted for material design
and testing ages [47, p.19]. In line with this, the Austrian standard ÖNORM B4452
requires Plastic Concrete to achieve an unconfined compressive strength (UCS)
of at least 0.5 MPa at 90 days, or alternatively an UCS of 0.3 MPa at 7 days
in the event of a water table draw-down within the first 90 days [108, p.10]. It
should also be noted that Plastic Concrete strength is generally higher than that
of cement-bentonite mixtures where a minimum compressive strength of 0.1 MPa
at 28 days is requested by some guidelines [86]. Cement-bentonite projects have
also been known to achieve a compressive strength of 0.5 to 1.5 MPa [55, p.39].
In practical Plastic Concrete applications the compressive strength has been
required to range between 1.0 and 2.0 MPa. Most recently, the Bagatelle Dam
Plastic Concrete cut-off wall required a compressive strength of 1.0 to 1.5 MPa at
28 days [25, p.39]. The Hinze Dam Plastic Concrete cut-off wall was expected to
achieve compressive strengths of 2.0 to 4.0 MPa at 28 days [24] [31, p.310]. It
should however always be taken into account, that due to the interdependance be-
tween concrete compressive strength and elastic modulus (which will be adressed
further on in this report), the required compressive strength should be considered
a target strength and not a minimal strength [95].
Deformation Requirements
ICOLD Bulletin No. 51 recommends the elastic modulus of Plastic Concrete to be

four to five times greater than that of the surrounding soil [87, p.11]. This aims to
achieve a material presenting similar deformation characteristics to the surround-
ing soil [87, p.11], thus decreasing the relative settlement between cut-off wall
and surrounding soils hereby reducing the so called arching effect [100]. This in
turn reduces the stress applied on cut-off wall material [77]. The Bagatelle Dam
project required for example a deformation modulus (with geotechnical testing
standards) of 100 to 150 MPa at 28 days [25, p.39]. The testing conditions for the
5
deformation modulus (e.g. testing age, strain level, drainage conditions, confin-
ing pressure & strain rate) are however not often standardised [86, p.N15]. Some
guidelines also require a specific strain at failure value to be achieved [86, p.N14].
Hydraulic Conductivity Requirements
Regarding the hydraulic conductivity of Plastic Concrete, most standards and

guidelines require similar values to be achieved. The DWA M512-1 guideline
states that with a hydraulic conductivity k3 ≤ 1 · 10−8 m/s, a hydraulic gradient of
i = 100 may be safely absorbed by a Plastic Concrete cut-off wall. For cement-
bentonite cut-off walls the British Institution of Civil Engineers recommends a tar-
get permeability of k = 1 · 10−9 m/s [86, p.S5]. Interestingly however, due to the
inherent variability within the material this guideline suggests that 80 % of speci-
mens have a permeability of less than 1 · 10−9 m/s, 95% of less than 1 · 10−8 m/s
and no single values less than 5 · 10−8 m/s, when testing at an age of 90 days or
later [86, p.S5].
In practical applications the permeability requirements have more recently
moved from 1 · 10−8 m/s to 1 · 10−9 m/s [86, p.N13]. The permeability of the Hinze
Dam (Australia) was required to be k ≤ 1 · 10−9 m/s [31, p.310] [24, p.47] and
in the Sylvensteindamm (Germany) a permeability k ≤ 1 · 10−9 m/s was also ex-
pected [95]. Since permeability is more sensitive to testing age and time under
permeation, with tighter specification requirements having been established, test-
ing has generally been recommended to move to at least 90 days [86, p.N13].
For example, the Austrian standard OENORM B 4452 requires Plastic Concrete
permeability testing to be performed up to 90 days of age [108, p.9].
1.5 Problem Defintion

Current design of Plastic Concrete is however simplistic with a linear-elastic ma-
terial model being used for material modelling. For cut-off wall and dam safety
it is therefore of utmost importance that Plastic Concrete is realistically designed
and sufficiently understood [29, 30]. Most importantly neither the time-dependant
properties (e.g. creep behaviour) nor the deformation properties (e.g. high duc-
tility) are considered when designing cut-off wall materials [30]. To date, there
are major uncertainties in the design and construction of all types of cut-off walls,
which must be addressed [125, p.8].
The following report therefore aims to review the State-of-the-Art on Plastic
Concrete for cut-off walls and recommend an approach for the correct modelling
of Plastic Concrete behaviour. In chapter 2 the raw materials, the mixture com-
6
position and the mixing sequence for Plastic Concrete are set out. In this chapter
the fresh properties of Plastic Concrete are also highlighted. In chapter 3 the
mechanical behaviour of hardened concrete is described in detail. Hereby the
different material testing standards for Plastic Concrete strength, elastic modulus
and creep behaviour are explained against the available literature. In chapter 4
the hydraulic behaviour of Plastic Concrete is described. Finally, in chapter 5 the
main results of Plastic Concrete behaviour are briefly summarised and the needs
for future research are proposed.
7
2. Mix Design
Contemporary standard concrete is considered a five-phase construction material

composed of cement, water, aggregate (sand and gravel), additions (e.g. sup-
plementary cementitious materials) and admixtures (e.g. set-retardants, super-
plasticisers, stabilisers). Plastic Concrete can also be considered a five-phase
construction material, however in differing proportions to those usually mixed with
standard concretes and containing bentonite as an additional constituent. In sec-
tion 2.1 the various materials used in the mix design of Plastic Concrete are men-
tioned and their specific properties and purposes are detailed. Following on, in
section 2.2 the mixture composition of Plastic Concrete is described and placed
within concrete technology context. In addition, in section 2.3 the existing mixing
sequence possibilities are described and weighed out against one another. Fi-
nally, in section 2.4 the testing requirements of Plastic Concrete fresh properties
are given.
2.1 Materials
As mentioned previously, Plastic Concrete can be considered a five-phase con-
struction material. Here cement, water, aggregate (mainly sand and fine gravel)
are used in combination with bentonite as an additive and sometimes admixtures
to obtain a highly ductile and impermeable material. However, the composition of
Plastic Concrete is not limited to the aforementioned components and could be
produced using other supplementary cementitious materials (e.g. fly ash). In the
following sections the most often used materials are described.
2.1.1 Cement
Cement is a type of binder with adhesive and cohesive properties commonly used
in the construction industry capable of binding building materials together. Hy-
draulic cements are most commonly produced by mixing calcareous and argilla-
ceous materials together and burning them at high temperatures (i.e. clinkering
temperature) and grinding the resulting clinker into a powder [105, p.2]. In the
8
PLASTIC CONCRETE FOR CUT-OFF WALLS MIX DESIGN
presence of water this dry powder undergoes a chemical reaction becoming adhe-
sive and forming calcium silicate hydrate (C-S-H) crystals which consolidate and
strengthen the structure. The variety of hydraulic cements available on the mar-
ket is manifold, as the use of cement is widespread in the construction industry
with many applications (e.g. ordinary Portland cement, sulfate-resisting cement,
blastfurnace cement). Their dry density therefore also range from 2.85 g/cm3
to 3.5 g/cm3 . In Europe the varying types of cement are regulated within a sin-
gle CEN-standard, which in Germany corresponds to DIN EN 197-1 [48]. Most
notably, the European standard classifies the various cements by composition,
whereby these are divided in five categories from CEM I (Portland cement) to
CEM V (composite cement). In the Unites States various standard apply to de-
scribe the varying type of cements. Firstly ASTM C150/C150M [7] regulates or-
dinary Portland cement, while ASTM C595/595M [11] standardises blended hy-
draulic cements. Some physical properties are also regulated separately in the
performance specification ASTM C1157/C1157M [6]. This should especially be
taken into account since an ASTM C150/C150M [7] Type I cement corresponds
to a DIN EN 197-1 [48] CEM I class cement. However, a DIN EN 197-1 [48]
CEM III/A class cement does not correspond to a ASTM C150/C150M [7] Type III
cement, but instead a Type I (36<S<65) cement. For more information regarding
cement classification and terminology, refer to [8, 105].
For Plastic Concrete two main choices exist. The International Commission
on Large Dams (ICOLD) recommends within its Bulletin 51 [87] the use of blast-
furnace (BLF) or pozzolan (POZ) cement since these types of cement have a
stronger resistance against chemically aggressive water, as is also common knowl-
edge within concrete technology [79, 105, 129]. In concrete construction it is also
known, that through the use of BLF cement, concrete strength development at
early age is much slower than with ordinary Portland cement [33, 79, 105, 114].
With this, when the secondary slurry-trench element is cut between two previ-
ously tremie-placed primary elements, the tremie concrete is still of low strength.
This in turn makes it possible for trench cutters to cut the secondary elements
precisely and with low wear for the cutter heads. However, the slow concrete
strength development can also be counter productive if not controlled, since very
low concrete strengths may cause construction operations to be halted before the
secondary element can be cut. In this case, other blends or a low strength or-
dinary Portland cement can be used to achieve similar compressive strengths at
28 days but higher strength within the first days, irrespective of Plastic Concrete
permeability. In addition, the regional availability of BLF or POZ cements may
also be a limiting factor when choosing the cement type to be used.
9
Some studies on the structure of single-phase diaphragm wall materials (such

as cement-bentonite mixtures) have however also shown that there is a differ-
ence within the material structure and mechanical parameters depending on the
cement type used [55, p.22]. When using ordinary Portland cement the struc-
ture is more open than with blast-furnace cement, which is identifiable by SEM
imagery shown in Figure 2.1 [55, p.24ff.].
(a) Ordinary Portland cement (b) Blast-furnace cement
Figure 2.1: SEM images of the micro-structure of hardened cement-bentonite mixtures

depending on cement type used [55]
These images also show that, whilst the cement particle hydration occurs sim-
ilarly in cement-bentonite mixtures and standard concrete, the mean particle dis-
tance increases from 2 µm for standard concrete to 15 µm for cement-bentonite
mixtures [55, p.26]. The authors ascribe this to the increased w/c-ratio, but also
to the presence of bentonite particles within the cement particle gap. They also
hypothesise that formation of C-S-H occurs differently depending on the cement
type used; with ordinary Portland cement, the high concentration of Ca2+ -ions at
the cement particles causes C-S-H to precipitate at the clinker particles. On the
the other hand, in blast-furnace cement the presence of slag particles further in-
creases the cement particle distance, causing the Ca2+ -ions to be more evenly
distributed. The authors therefore hypothesise that C-S-H may also form at the
bentonite platelets, causing the platelets to be adhered together [55, p.27]. Test
results showed hereby that the cement-bentonite strength increases and perme-
ability decreases, with this effect being further reinforced the higher the slag con-
tent is [55, p.19ff.]. However, to date, this hypothesis remains unconfirmed. In ad-
dition, the interaction between cement and bentonite particles in Plastic Concrete,
i.e. in the presence of aggregates and further admixtures, remain unexplored and
should be subjected to further study.
10
2.1.2 Bentonite
Bentonite is a weathered rock composed of clay-like minerals which was first
discovered in 1898 in Fort Benton, MT (U.S.A.) and is an alteration product of vol-
canic ash [78]. Although the bentonite discovered in Fort Benton is mainly com-
posed of montmorillonite minerals (≥ 80 wt-%), the term bentonite is however now
well established and encompasses any clay-rock composed of smectite minerals,
which in turn dominate the physical properties of the rock [78, p.1f]. The oven-dry
density of bentonite generally ranges between ρ ≈ 2.65 - 2.75 g/cm3 [124, p.276].
Structure & Properties
Smectite minerals form platelets composed of three layers. The most common
smectite mineral, montmorillonite, consists of two SiO4 -tetrahedron on opposite
sides of a AlO6 -octahedron [112, 128]. Due to the partial, isomorphic substitution
of some cations a layer charge is generated. This negative layer charge is in turn
counter-balanced by other cations within the interlayer space between two adja-
cent platelets. Most commonly the interlayer cations are Ca2+ , Mg2+ or Na+ which
neutralise the negative surface charge, and account for the two main bentonite
groups Na-bentonite and Ca-bentonite (which commonly includes magnesium-
bentonites) [90] [97, p.175f.] [112, p.10]. These interlayer cations do not however
form ionic bonds but are instead bound through van-der-Waals interactions [106,
p.1992]. In Figure 2.2 a schematic illustration of the montmorillonite structure is
given, whereby the SiO4 -tetrahedrons and AlO6 -octahedron as well as interlayer
cations (red) can be seen.
Figure 2.2: Schematic illustration of montmorillonite structure [112]
Furthermore, the weak layer charge permits the interlayer cations to adsorb
and retain water molecules [85,97]. The water adsorption capacity of sodium and
calcium bentonite is however disparate, with Ca-bentonite adsorbing 200-300%
water, while Na-bentonite can adsorb up to 600-700% of water [76, 112]. More-
11
over, water adsorption does not occur instantaneously. Ca-bentonite adsorbs

most water within the first minutes [78, 85]. On the other hand, Na-bentonite
adsorbs water more slowly, whereby water adsorption (also called hydration in
the support fluid industry) is still not completed after 18 h [78, p.241]. Water
is not however solely bound around interlayer cations, but water molecules also
adsorb on to the negatively charged surface of the SiO4 -tetrahedrons [96]. The
adsorbed water is hereby submitted to high surface tension (≥ 2000 MN/m2 ) and
is strongly hindered from any movement [96, p.216]. The surface tension of ad-
sorbed water, together with the low particle size and hydrated interlayer cations,
causes bentonite and bentonite-composites to have a low permeability, since the
water is hindered from transportation by the corresponding bonds [96, p.217].
The aforementioned water adsorption phenomena cause the clay minerals, es-
pecially montmorillonite to significantly increase in volume, multiplying its starting
volume manifold. Due to the differing water adsorption capacity, Ca-bentonites
exhibit a smaller swelling potential with the interlayer gap increasing up to a max-
imum value of 2 nm [85, p.I-6]. Na-bentonites on the contrary may swell far more
strongly with the interlayer water even separating the individual platelets from one
another [85, p.I-7].
Most naturally occurring bentonites predominantly have calcium ions in the
interlayer space and are far more abundant than the more active Na-bentonites.
However, Ca-bentonites may be "activated" to obtain Na-bentonites by exchang-
ing Ca-cations with Na-cations within the interlayer space, enhancing material
performance [97, p.176]. This most naturally occurs by mixing soda ash (Na2 CO3 )
to Ca-bentonites with the partial precipitation of calcium carbonate (CaCO3 ), al-
though other technical procedures exist [90, 112]. This process is called alkaline
activation, and is thoroughly described in Jasmund & Lagaly [90, p.363f.]. The
activated Na-bentonites also have a more constant product quality than natural
Na-bentonites due to this procedure [112, p.20] and may also be economically
obtained with the appropriate process technology [106]. This cation exchange is
however almost completely reversible and may occur when the bentonite is put
in contact with the released Ca2+ -ions from cement [97, p.181], causing cement-
bentonite mixtures to flocculate and become unstable [85]. Therefore for cement-
bentonite applications a cement-stable bentonite is required, which maintains its
rheological, swelling and mechanical properties in the presence of cement parti-
cles, allowing for the production of stable cement-bentonite mixtures [97, 107].
Although some contemporary, commercially available bentonites have proven
compatible with cement in some applications, the reasons for bentonite cement-
stability are not yet fully understood. Some authors suggest the content of free
soda-soluble silica or the presence of accessory minerals in bentonite may be
12
a determining parameter for cement-stability [107]. It is therefore imperative to

further study the cement-bentonite interaction to establish the reigning interaction
mechanisms [84].
Characterisation
At present, no satisfactory specifications for bentonite for slurry trench works ex-
ist [86, 93]. Solely the OCMA specification DFCP-4 for drilling fluid material ben-
tonites is sometimes cited, does however not fulfill slurry trench requirements [86].
For this reason, standard geotechnical testing procedures are commonly used to
characterise commercially available bentonites [92, 93]. Bentonite as a mineral or
soil, can therefore be characterised using the Atterberg limits, whith testing fol-
lowing DIN 18122-1 [57] or ASTM D4318 [17]. The Atterberg limits are a basic
measure of the water content of a fine-grained soil. The limits most commonly
measured are the plastic limit (wp ) and liquid limit (wl ). The difference of these
two limits, is the so called plasticity index (Ip ) and defines the range size of water
contents where the soil exhibits plastic properties [32, p.107f.]. In Figure 2.3 an
overview of the soil states and corresponding water contents is given. For ex-
ample, regular silt has a plasticity index Ip in the range of 10, whilst clay usually
lies at around 30 [117, p.54]. On the contrary bentonite commonly has a plasticity
index of 100 with a liquid limit close to 150, far exceeding regular soils [78, p.239].
Ip = wL - wp
solid semi-solid plastic liquid
w
0 ws wp wL 1
ws = shrinkage limit wp = plastic limit wL = liquid limit
Figure 2.3: Overview of soil states and corresponding water content
Another parameter is the so called activity index (IA ), which is defined as the
ratio of plasticity index (IP ) to the percentage of clay-size particles (i.e. particles
with d < 0.002 mm). In Equation 2.1 the activity is given. The activity of a given
soil may shed light on the minerals contained within the soil [74, p.357f.] [117].
Regular kaolinite has an activity index of approximately 0.4, Ca-bentonite of ap-
proximately 1.5 and Na-bentonite often reaches values greater than 7 [78, 117].
The presence of clay minerals can be confirmed with X-ray diffractometry.
In addition, the amount of clay present in aggregate can be determined using
the methylene-blue test, which also measures the cation-exchange capacity of
soils [111, p.290]. Some authors also recommend using the methylene-blue test
to determine the montmorillonite content within a bentonite sample, since this is
also decisive for the bentonite properties [55, p.11].
13
IP
IA = (2.1)
mT /md
where IA : activity index (-)

IP : plasticity index (-)
mT : dry mass with d < 0.002 mm (g)
md : dry mass with d < 0.4 mm (g)
Furthermore within the context of specialist foundation construction it may be

advisable to test the bentonite swell index as well as the corresponding betonitic
slurry free-fluid fraction. The former is tested following ASTM D5890 [19] within
laboratory conditions and enables the evaluation of swelling properties of any clay
mineral which may be used for hydraulic conductivity reduction. The free-fluid
volume fraction is determined following DIN EN ISO 10426-2 [51] and is useful to
understand the static stability of the bentonite or cement-bentonite slurries.
Last but not least the water adsorption capacity of bentonite is most commonly
measured following DIN 18132 [60] for soils and DIN EN ISO 10769 [52] specifi-
cally for bentonite. In the latter, the previously dried bentonite specimen is placed
on a filter and the adsorbed water is measured over a defined period of time.
Application
Due to the previously mentioned structure, bentonite is highly swellable and thixot-
ropic and has therefore manifold industrial applications especially in the construc-
tion industry. Most commonly bentonite has been used in the specialist founda-
tion construction when placing slurry trench walls, where a bentonite-suspension
is used to hydrostatically stabilise the adjacent ground [124]. Furthermore, ben-
tonite is used as a stabilising agent in drilling fluids and cement suspensions
improving workability and reducing the risk of segregation [78, 103, 113]. Some
authors also use bentonite to waterproof soil and structures, reducing the overall
permeability [74, 78, 113]. Finally, bentonite is also used in regular and nuclear
waste disposal, since the high cation exchange capacity of montmorillonite en-
ables bentonite to adsorb chemical pollutants and heavy metals, hindering their
passage through a seepage barrier [85, 96].
Bentonite has historically been used in Plastic Concrete as it was a commonly
available, cheap stabilising agent for Plastic Concrete mixtures [73, 76]. In addi-
tion bentonite allows for a more ductile behaviour of a Plastic Concrete diaphragm
wall [74, 87, 89, 125]. Some authors suggest that through bentonite swelling the
air voids within the cement paste structure are filled partially and the permeabil-
ity is reduced [74, 76, 118]. This however only likely happens if the bentonite
14
has not previously been completely hydrated, allowing for a further bentonite hy-
dration in the hardening Plastic Concrete. However, the mixing sequences in
Plastic Concrete production are manifold and heavily depend on construction site
specifications, hence a full or partial hydration can not be safely assumed (see
section 2.3).
Scholz et al. studied the mechanical properties of various diaphragm wall
compounds and stated that the average pore size within Na-bentonite specimens
is notably smaller than those within Ca-bentonite specimens [118]. In addition,
the bentonite structure varies, with sodium-bentonite presenting a face-to-face
structure, whilst calcium-bentonite shows a combined face-to-face, edge-to-edge
structure. Calcium-betonite has gel pores in addition to capillary pores [118]. This
in turn affects the permeability of bentonite-mixed materials, since for example
small amounts of bentonite significantly reduce the permeability of sands. Na-
bentonite mixed sands have a far lower permeability than Ca-bentonite mixed
sands with difference being two orders of magnitude [78, p.243].
2.1.3 Aggregates and Admixtures

The most important criteria for the choice of aggregates in Plastic Concrete is the
maximum grain size, due to the high segregation risk of fresh Plastic Concrete.
This is caused by the relatively high w/c-ratio and the need to use bentonite as
a stabilising agent (see section 2.2). Therefore, aggregates are generally limited
to sands and fine to medium gravels. According to the DWA guideline M512-1
a maximum grain size of dmax = 63 mm should be adopted [56, p.60]. Similarly
ICOLD Bulletin 51 recommends a maximum grain size of dmax = 30 mm and does
not contain too large a fraction of fines [87, p.29]. Practical examples and ex-
perience have however shown that such mixtures show very strong segregation
effects, and therefore smaller grain sizes should be used. Most commonly a
maximum grain size of dmax ≤ 12 mm is used for Plastic Concrete mixtures. The
USBR Design Standard 13(16), recommends maximum grain size to be limited
to dmax ≤ 25 mm [125, p.59]. The Austrian standard ÖNORM B4452 [108] limits
the maximum grain size to dmax ≤ 22 mm, however notes that maximum grain
size above 16 mm is rare. The standard also states that special consideration
should be given to the segregation risk and deformability of Plastic Concrete mix-
tures when using a maximum grain size dmax ≥ 8 mm. Furthermore, the fine
particle content is also partially regulated to guarantee the necessary flowabil-
ity [108, p.4]. It should however be noted that it is often difficult to meet specific
grading demands at building sites in some countries. Furthermore, rounded ag-
gregate is preferred as this type of aggregate further enhances the flowability
15
of tremie concrete [130]. Moreover, the type of aggregate used is regulated by

the exposure to any aggressive contaminant, with quartz based aggregate be-
ing the preferred aggregate type [130]. Occasionally, additional mineral fillers
(i.e. stone dust, clay dust and fly ash) are added to Plastic Concrete mixtures to
further increase the solids content, although this is not a commonly used proce-
dure [85, p.VIII-9].
Various types of admixtures are also used in Plastic Concrete mix designs.
Most often, retarding admixtures are used to slow down concrete setting and
prevent premature concrete stiffening [125, p.57]. With this a longer workability
window is achieved and longer slurry trench elements can be produced, for which
concrete placement with the tremie method can be safely finalised. Depending
on Plastic Concrete mixture composition varying amounts of retarding admixtures
may be added normally ranging from 1 wt-% to 2.5 wt-% of cement content [24,
p.47]. Especially with long slurry-trench elements, the retarding agent dosage
has to be carefully measured, since a shortfall of retarding agent can cause the
first concrete batches to stiffen and then in turn be displaced like a plug above the
fresh tremie placed concrete [24, p.47].
In some cases superplasticizing admixtures are also used to ensure better and
more controlled workability of the Plastic Concrete mixture. It should however be
noted that the effectiveness of modern polycarboxylate ether-based superplasti-
cizers (PCEs) is negatively affected by the presence of clay minerals, especially
montmorillonite [111, p.289]. Therefore, corrective actions should be taken when
using PCEs (e.g. the use of PCEs with hydroxyalkyl side chains) [111, p.290]. Ex-
ceptionally further stabilising admixtures are used to stabilise the fresh concrete
if the bentonite proves to be insufficient.
In most instances tap water is generally suitable for Plastic Concrete produc-
tion. However, untreated water or water with high ion concentrations may affect
bentonite dispersion or hydration process and should therefore be tested in trial
mixtures if required [86, p.N19].
2.2 Mixture Composition

As mentioned previously, Plastic Concrete is also a five-phase material. In con-
trast to standard concrete, the w/c-ratio of Plastic Concrete is much higher, with
values ranging from 3.3 to 10 [87]. The cement content is also significantly lower
than that of standard concrete, rarely surpassing the 200 kg/m3 mark and even
being as low as 80 kg/m3 . In addition, Plastic Concrete contains bentonite with the
bentonite content laying anywhere between 0.5% and 12% by weight of dry mass
of constituents depending on the mixture composition and bentonite type. Gen-
16
erally speaking, the lower bentonite contents correspond to Na-bentonite, whilst

the higher contents correspond to Ca-bentonite, which is caused by the differ-
ing swelling behaviour. Some authors suggest a Na-bentonite to Ca-bentonite
equivalence of 1 : 5, suggesting a more economic approach through the use
of Na-bentonite [55, p.15]. Furthermore, Plastic Concrete contains somewhat
smaller similar quantities of aggregate to standard concrete, hereby ranging from
1100 kg/m3 to 1500 kg/m3 .
In Figure 2.4 five different concrete mixtures are shown, of which three corre-
spond to Plastic Concrete mixtures. The standard concrete example from Grübl et
al. [79] represents a standard concrete with 20 MPa strength at 28 days. The mid-
dle three mixtures are examples for Plastic Concrete with an approximate com-
pressive strength of 1.30 MPa at 28 days [22, 116]. Finally, a mixture composition
by Triantafyllidis [124] of a single-phase diaphragm wall material with 1 MPa com-
pressive strength at 28 days is given for comparison.
2 5 0 0
G ra v e l
S a n d
C e m e n t
2 0 0 0 W a te r
8 6 2 .6 B e n to n ite
9 3 7 .5 4 4 2
M a s s in k g /m ³
7 4 2 .2
1 5 0 0
1 0 3 2 2 2 5
1 0 0 0 9 5 3 .4
6 0 7 .4 9 3 7 .5
ω= 4 ,1 0
5 0 0 1 8 0 8 0 9 1 5
3 1 7 ω= 2 ,2 1 1 2 0 ω= 4 ,9 4
ω= 0 ,6 0 3 9 5
3 9 7 ω= 2 ,0 2
2 4 2
1 9 0
0
G rü b l B a g h e ri S a d r e k a r im i B A U E R T r ia n ta fy llid is
2 0 0 2 2 0 0 8 2 0 0 2 2 0 0 4
fc u = 2 0 M P a fc u = 1 ,2 7 M P a f c u ,c u b e = 1 ,2 6 M P a f c u ,c u b e = 1 ,3 3 M P a fc u = 1 M P a
S ta n d a rd P la s tic C o n c r e te S in g le P h a s e
C o n c re te D W M
Figure 2.4: Representative examples of Plastic Concrete mix designs
As can be seen, the Plastic Concrete mix design is a combination of standard

concrete and single-phase diaphragm wall material. The use of aggregates (most
notably sand and fine gravel) in somewhat reduced quantities compares to the
composition of standard concrete. The density of Plastic Concrete is also similar
to that of concrete ranging from 1.9 g/cm3 to 2.3 g/cm3 . The Austrian standard
ÖNORM B4452 requires a minimum density of 1.80 g/cm3 [108, p.9], since a
density difference of 0.5 g/cm3 is necessary to effectively displace the bentonite
slurry within the slurry trench element without mixing, when placing concrete with
17
the tremie method [108, p.14] [124]. Other authors recommend a density differ-
ence of 0.75 g/cm3 [124, p.270]. The w/c-ratio on the other hand compares to
that of single-phase diaphragm wall materials, exceeding by far 1.0 implying the
existence of a far coarser micro-structure. Also, the use of bentonite as a stabilis-
ing admixture is comparable to that of single-phase diaphragm wall materials. It
should be emphasised that single-phase diaphragm wall materials are not placed
with the tremie method and therefore is also subjected to other construction un-
certainties and limitations [74, 124].
2.3 Mixing Sequence

Across the literature the Plastic Concrete mixing process is not consistent. Vari-
ous options are presented, which are schematically shown in Figure 2.5.
Figure 2.5: Representative examples of Plastic Concrete mixing sequences
Alternative A is the most commonly described variant in literature [22, 28, 72,
80, 89, 101, 110, 116]. In this bentonite and water are gradually mixed together
and the let to hydrate for up to 24 h. After this, cement is added to the bentonite-
suspension and thereafter the aggregates are added. Alternative B [135] pre-
mixes the components bentonite, cement and aggregates to a dry compound.
This compound is then mixed with water and placed within the slurry-trench ele-
ment without allowing for any hydration time. This method is however not com-
monly used in practice, since the dry mixing of components and subsequent water
addition does not achieve a sufficiently homogeneous Plastic Concrete mix [95,
p.240]. Finally, alternative C is an often used mixing sequence by some construc-
tion companies, whereby cement and aggregate are mixed to a dry compound,
whilst bentonite and water are mixed into a slurry [95]. The bentonite slurry is then
mixed with the dry compound to obtain the Plastic Concrete mixture, whereby the
bentonite slurry is not allowed to hydrate before use. Alternative C can therefore
be considered a combination of alternative A and B. It should be noted however
18
that due to the differing mixing sequence and hydration time, varying results can
be expected in terms of mechanical properties and permeability values. This is
most likely the fact, as the bentonite hydration phase is different for the three
aforementioned alternatives, which in turn affects the void filling in the hardened
cement paste.
The hydration of bentonite is however not only dependant on the aforemen-
tioned differences between bentonite types (see subsection 2.1.2) but also on
the type of mixer and thus the induced shear rate γ̇. For any given mixer it can
be seen that the higher the maximum achievable shear rate γ̇ is, the shorter the
hydration time required for bentonite samples will be [55].
2.4 Fresh Properties

To ensure the correct placement of concrete, which in turn enhances hardened
concrete quality, the fresh properties of Plastic Concrete mixtures have to be con-
trolled, especially concrete flowability during the whole casting process. There-
fore, the fresh properties must be controlled not only during inital placement,
but also measure the thixotropic and flow retention characterestics of the con-
crete [98, p.215f.]. The flow properties are mainly affected by concrete rheology
which in turn result from the concrete mix design [98]. Despite the complexity and
relevance of concrete rheology it is still not uncommon for simple concrete testing
procedures to be used to determine the fresh properties of concrete [98, p.211].
It should be noted that many problems in diaphragm walls may be attributed to
the use of ianadequate concrete mixes resulting from poor concrete specifications
due to deficient or simplistic testing procedures [70, 98]. Most commonly the so
called slump test [38] and flow table test [39] are used, although other tests do
exist. In Table 2.1 an overview of the most common testing procedures for fresh
concrete flowability and their corresponding testing standard is given.
Table 2.1: Overview of common testing procedures for fresh concrete workability
German DIN-Standard English ASTM-Standard

Setzmaß EN 12350-2 slump C143/C143M
Setzfließmaß EN 12350-8 slump-flow C1611/C1611M
Ausbreitmaß EN 12350-5 flow table -
L-Kasten EN 12350-10 L-Box -
For Plastic Concrete placed with the tremie method various guidelines and
standards exist, which require specific values of concrete fresh properties. Ac-
19
cording to DIN EN 1538 [47] which in turn refers to Appendix D of DIN EN 206 [50]
the flow table test values for concrete placed with the tremie methods should be
600 mm. Alternatively, a slump test can be performed with a target value of
200 mm. Similarly, the Austrian standard ÖNORM B4452 [108] requires the con-
crete to obtain flow table test values in the range of 55 cm to 65 cm. This standard
also limits the free-fluid test value following DIN EN ISO 10426-2 [51], commonly
used to test sedimentation stability, to 2% after 2 hours [108, p.9]. The DWA
guideline M512-1 [56] recommends a flow table test value greater than 530 mm
for Plastic Concrete. Similarly, in USBR Design Standard 13(16) a 15 cm to 23 cm
slump is desired to ensure a high degree of fluidity and workability [125, p.56].
Furthermore, the density should be measured following EN 12350-6 [40] when
the Plastic Concrete dosing is volumetric.
Other tests such as for example the L-Box test following DIN EN 12350-10 [37],
although developed for super-workable concrete, is not entirely adequate for Plas-
tic Concrete since the high flowability and low maximum aggregate size makes
the containment of the concrete within the L-Box difficult. This in turn does not
allow for the calculation of the passing ability ratio (PL). For a more detailed in-
vestigation into the application of the L-Box test for tremie pipe concrete refer-
ence is made to [1]. Some guidelines also refer to the Marsh funnel viscosity
following DIN 4127 [66] or ASTM D6910 [20], when determining the fresh con-
crete flowability. It should be noted however that a Marsh funnel has a maximum
opening at the bottom of 4.75 mm and a entry screen of 3.2 mm, which in turn
only really allows for the measurement of bentonite-slurries without aggregate.
Nonetheless, ICOLD Bulletin 51 requires a Marsh funnel viscosity of 50 s for the
bentonitic slurry [87, p.27]. Other testing methods for Concrete workability for
Deep Foundations exist (e.g. Modified Cone Outflow test) and may be reviewed
in [71, Apx.A].
Concrete flowability is generally controlled through the water content and su-
perplasticizing agent content, however the stability of the Plastic Concrete have
to be closely monitored. Evans et al. [72] also suggest, that the workability of
of Plastic Concrete is enhanced through the addition of fly ash, due to its ball
bearing type action.
For more detailed information regarding the various fresh concrete testing
methods applicable for tremie method refer to the EFFC/DFI Guide to Tremie
Concrete for Deep Foundations [71]. Scientific fundamentals on concrete rheol-
ogy can be found in [115].
20
3. Mechanical Behaviour
Especially important is the precise description of the mechanical behaviour of

Plastic Concrete. For this various testing methods and standards exist to study
the mechanical properties, which are described in section 3.1. In section 3.2 the
mechanical strength of Plastic Concrete using the aforementioned testing meth-
ods is described. The deformation properties of Plastic Concrete (e.g. Young’s
modulus) are set out in section 3.3. Finally, the relaxation potential of Plastic
Concrete is reported in section 3.4.
3.1 Testing Standards

Plastic Concrete has a relatively low compressive strength at 28 days in the range
of 1 MPa to 3 MPa. This in turn causes Plastic Concrete to have a very low
compressive strength for concrete testing methods, but a rather high compressive
strength for soil testing methods. Plastic Concrete can therefore be considered to
be in the transition zone of these testing methods.
3.1.1 Compressive Strength

The most common testing methods for concrete strength is the unconfined com-
pressive stress (UCS) test. In this concrete samples are placed within as testing
machine and axially loaded. The standard European test method is EN 12390-
3 [44], the corresponding ASTM Standard is ASTM C39/39M [9].
For the testing of soil strength various testing methods are used. Firstly, the
UCS is measured in a similar manner to that of concrete, following DIN 18136 [62]
or ASTM D2166/2166M [14]. However, for soil various other shear strength and
triaxial compressive strength test methods exist. Triaxial testing is herein di-
vided into three types of triaxial tests; Consolidated Drained (CD), Consolidated
Undrained (CU), Unconsolidated Undrained (UU). The main diferences between
the three types, lie within the consolidation state of specimens and the build up
of pore water pressure. Consolidated specimens are previously loaded until the
measured total stress σ equals the measured effective stress σ0 , i.e. that the pore
21
PLASTIC CONCRETE FOR CUT-OFF WALLS MECHANICAL BEHAVIOUR
water pressure u = 0, as shown in Equation 3.1. In drained experiments water

can gradually exit the specimen and therefore no pore pressure build up exists
(∆u = 0). In undrained experiments the system is "closed" with which the pore
pressure increases (∆u 6= 0) and is measured during CU tests [117]. It should
be however noted that for the practical measurement of Plastic Concrete the pore
water pressure is of negligible influence.
σ0 = σ − u (3.1)
where σ0 : effective stress (MPa)

σ : total stress (MPa)
u : pore water pressure (MPa)
In Germany all triaxial tests summarised in one standard and performed fol-
lowing the DIN 18137 testing standard [63, 64]. The ASTM standards divide the
three testing methods CD, CU and UU into the three standards ASTM D7181-
11 [21], ASTM D4767-11 [18] and ASTM D2850-15 [16], respectively.
It is also interesting to note that geotechnical standards solely test cylindri-
cal samples with varying height-to-diameter (h/d) ratios. Concrete standards
allow for compressive strength testing on cylindrical samples with h/d = 2, but
also on cubic samples which is of common practice. This should be consid-
ered, since both the h/d-ratio as well as the specimen shape affect the test
results obtained, but no direct conversion formulae exist [105, p.596]. In con-
crete technology it is of common knowledge that cubic specimens have a higher
strength than cylindrical samples, and an increasing h/d-ratio further decreases
the tested strength [79, 109, 114]. Cylinders are believed to give a greater unifor-
mity of results as their strength is less affected by coarse aggregate properties,
lesser end restraint influence and more uniform stress distribution in horizontal
planes [105, p.596].
Some authors also use other, non-destructive testing methods to estimate the
compressive strength. For cement treated soils, various authors have related the
compressive strength of cement-treated soils to the small strain shear modulus
measured with bender elements [119, 127]. However, these testing methods are
not standardised for concrete testing and can therefore only be considered for
compressive strength estimation.
3.1.2 Elastic Modulus

On the contrary, the Elastic Modulus E is defined and tested very differently de-
pending on the field of study. For concrete, the elastic modulus is defined as the
22
secant modulus, i.e. the slope between two points where Hooke’s law is appli-
cable. This modulus is of utmost importance, since it is typically used for anal-
ysis purposes [125, p.58]. Testing is performed following DIN EN 12390-13 [41]
or ASTM C469 [10], whereby the elastic modulus is determined between two
pre-set stress levels (σa = fc /3 and 0.10 · fc ≤ σb ≤ 0.15 · fc ). The neces-
sary stress is applied through the pressure plates of the testing machine while
strain is most commonly measured using strain gauges or linear variable differ-
ential transformers (LVDT). For more information regarding strain measurement
possibilities refer to [81]. In Equation 3.2 the elastic modulus definition following
DIN EN 12390-13 [41] is given.
∆σ σma − σb
m
EC,0 = = (3.2)
∆ε0 εa,1 − εb,0
where EC,0 : initial elastic modulus (MPa)

σm
a : measured upper testing stress (MPa)
m
σb : measured lower testing stress (MPa)
εa,1 : measured strain at the upper testing stress (-)
εb,0 : measured strain at the lower testing stress (-)
On the other hand, the testing of soil deformability has manifold possibilities.
The most common testing methods are the unconfined compression test and Oe-
dometer consolidation test. Within the unconfined compression test, the stan-
dard defines a deformation modulus as the "modulus of the uniaxial compression
test (Eu )" which is determined from the maximum tangential slope of the stress-
strain line following DIN 18136 [62]. DWA guideline M512-1 [56] for example rey-
commends the use of Eu as the defining parameter. ASTM D2166/2166M [14] on
the other hand uses the initial tangent modulus for this measurement. It should be
noted however that the strain measurement is not performed directly on the spec-
imen, but instead determined from the measurement of the piston movement.
This most likely also incurs in the measurement of machine displacement which
in-turn provides less accurate strain measurements of the Plastic Concrete spec-
imens [82]. Alternatively, soil deformability can also be measured with the Oe-
dometer consolidation test following DIN 18135 [61] or ASTM D2435/2435M [15].
In this, a specimen is compressed within a confining ring by imposing a load
over a frame in drained conditions (σ = σ0 ). The sample compression is then
subsequently measured over time by a dial indicator. With the resulting stress-
settlement curve the Oedometer modulus Es (German: Steifemodul) can be cal-
culated with Equation 3.3 [117, p.127ff.].
23
∆σ0
Es = (3.3)
∆ε
where Es : Oedometer modulus (MPa)
∆σ0 : effective stress increase (MPa)
∆ε : strain increase (MPa)
It should be noted however, that the Oedometer modulus is dependant on the

stress range used and is measured in confined conditions, unlike the aforemen-
tioned moduli E and Eu .
A tabular listing with DIN standards and corresponding ASTM standards for
the aforementioned testing methods is given in Table 3.1.
Table 3.1: DIN and ASTM Test Method Standards Comparison
Field Test DIN-Standard ASTM-Standard

Compressive strength (UCS) EN 12390-3 [44] C39/C39M [9]
Concrete Young’s modulus EN 12390-13 [41] C469/C469M [10]
Triaxial Tests - C801* [13]
Compressive strength (UCS) DIN 18136 [62] D2166/D2166M [14]
Oedometer consolidation DIN 18135 [61] D2435/D2435M [15]
Soil Triaxial Tests - CD DIN 18137 [63, 64] D7181 [21]
Triaxial Tests - CU DIN 18137 [63, 64] D4767 [18]
Triaxial Tests - UU DIN 18137 [63, 64] D2850 [16]
* = standard withdrawn
All in all it should be recommended to use concrete testing standards, since

Plastic Concrete composition (see section 2.2) and properties correspond to those
of concrete materials. In addition, the direct strain measurement of samples
(e.g. with strain gauges, LVDTs, etc.) is much more precise than machine dis-
placement measurements. However, none of the aforementioned standards is
ideal since Plastic Concrete’s strength is very low compared to standard concrete
and rather high for soil, which in turn requires an adjustment testing parameters
(e.g. loading speed, specimen preparation, strain measurement, etc.). Further-
more, the linear-elastic stress-strain relationship implied by Hooke’s law is also
not necessarily applicable for Plastic Concrete samples [92]. DWA M 512-1 [56]
guideline recommends for example that the UCS should be tested according to
DIN EN 12390-3 [44] for an expected UCS > 2 MPa and tested according to
DIN 18136 [62] for an expected UCS < 2 MPa. It should be however noted that
the real sample strength at a tested UCS of 2 MPa is not identical for both testing
methods. It is therefore recommended that the exact testing conditions should
24
therefore be specified during planning and tendering of projects to avoid testing

induced differences of acceptance criteria [25, p.41].
3.1.3 Testing Influences

It is of common knowledge in concrete testing, that various influencing factors
exist which affect testing results. Most notably the loading speed, but also sam-
ple preparation and storage conditions can affect the obtained test results. An
overview regarding the various influencing factors on concrete strength testing
can be found in [109]. In the following the effect of some of these influencing
factors are described in more detail against the background of Plastic Concrete
specimens.
Specimen end conditions
The ASTM standard practice for bonded capping of cylindrical concrete spec-
imens is ASTM C617 [12]. In Germany the capping process of concrete speci-
mens is embedded within the appendix A of DIN EN 12390-3 [44]. Herein four dif-
ferent methods are anchored namely sanding, calcium aluminate cement mortar
capping, sulphur mortar capping and the sandbox method. This standard states
that the application of calcium aluminate cement mortar and sulphur mortar is
limited to specimens with an expected strength of 50 MPa. This is due to the fact
that the mortar pastes have a relatively low elastic modulus and therefore deform
significantly when high loads are applied to test high strength concrete [53, 54].
Generally, capping materials should be at least as strong as the concrete they are
bonded to [109]. In addition the capping layer thickness should be kept to a min-
imum, as Dahm et al. [53, 54] showed that a greater layer thickness changes the
state stress within the specimen. The authors also note that cylindrical specimens
are affected less by specimen capping variations than cubic specimens, exhibit-
ing less scatter in the test results. For Plastic Concrete specimens it should be
however noted that these specimens have a very low strength. For this reason
specimen capping with mortar application should be unrestrictedly possible. In
addition the literature review shows that various capping methods are applied,
without any single specimen standing out [80, 93]. It should also be mentioned
that due to the high w/c-ratio there is a relatively high sedimentation potential. In
the event of sedimentation Plastic Concrete specimens exhibit a water-rich layer
in the upper specimen section, which in turn incurs in low concrete strength. It is
therefore necessary to account for possible sedimentation when reviewing Plastic
Concrete test data.
25
It is also questionable however, whether specimen capping is actually neces-

sary for Plastic Concrete samples. During the loading of concrete specimens with
uneven ends, critical stress peaks can occur which reduce the measured con-
crete strength. The high relaxation behaviour of Plastic Concrete however likely
incurs in the reduction of critical stress peaks. It remains to be shown, whether
specimen sanding or cut-off may affect Plastic Concrete structure by inducing
strength-reducing cracks and stresses.
Loading speed
It is common knowledge that concrete is a crack afflicted material [114]. Therefore

with increasing loading speed, the measured concrete strength increases as the
possibility of crack propagation around aggregate particles is reduced favouring
particle rupture [109, 114]. At very high rates of loading additional inertial effects
may occur [114]. At very low loading speeds, creep deformation may also occur
in addition to elastic deformation, causing concrete testing to determine lower
compressive strength [79, 109, 114]. Concrete has a permanent load resistance
(the so called creep strength) of 70% of the short-term resistance at a loading
speed of 0.2 MPa/s [105].
In DIN EN 12390-3 [44] the loading speed for concrete specimens is set to
0.6 ± 0.2 MPa/s. It should be noted that at this loading speed standard concrete
mixtures take 60 to 90 seconds to reach there peak load and rupture. The Na-
tional Annex of DIN EN 12390-3 also notes that at compressive strengths above
80 MPa or below 20 MPa the loading speed may be adjusted.
ASTM C39/C39M [9] establishes that a rate of movement corresponding to a
specimen stress rate of 0.25 ± 0.05 MPa/s should be applied. During the first
half of the anticipated loading phase, a higher rate of loading is permitted, the
concrete specimen should not however be subjected to shock loading.
On the other hand, geotechnical testing standards for soil DIN 18136 [62]
and ASTM D2166/D2166M [14] use strain rate as the defining loading speed.
DIN 18136 [62] establishes a strain rate of 1% of the sample height per minute
while ASTM D2166/D2166M [14] requires a strain rate between 0.5% and 2% of
the sample height per minute.
Neville however states that within the practical range of loading rates (0.07
to 0.7 MPa/s), the measured strength of standard concrete varies only between
97% and 103% of the strength obtained at 0.2 MPa/s [105]. Neville also notes
that stronger concrete exhibits lower sensitivity to the strain rate [105], which
may suggest higher loading speed sensitivity for Plastic Concrete specimens.
It should also be noted that standard concrete testing machines may be limited
in their measuring precision, as the application of axial loads lower 40 kN may
26
be subjected to high scatter or even impossible. Measurement accuracy should

therefore always be ensured before testing [45].
According to Kazemian et al. [93] the stress-strain behaviour of Plastic Con-
crete differs from that of ordinary concrete (not linear between 0% to 40%) and,
as expected, the standard loading speed is generally too high to measure stress-
strain. Hinchberger et al.’s study [80] with strain controlled experiments also
showed that Plastic Concrete is sensitive to compression rate, whereby higher
compression rates (0.01 mm/min > 0.001 mm/min) result in higher compressive
stress values, as would be expected [80, fig.7].
It is therefore necessary to adjust the standard loading speed in concrete stan-
dard for Plastic Concrete specimens to achieve measurable and precise data,
which is also in line with DIN EN 12390-3 [44] for specimens with compressive
strength below 20 MPa. Normally, and in accordance with normal strength con-
crete testing procedures, specimen failure should be achieved within 60 s to 90 s,
which means that a Plastic Concrete specimen with an expected 2 MPa strength,
should be tested with a loading speed between 0.02 MPa/s and 0.03 MPa/s. For
example, in DIN 4093 [65], which regulates the design of strengthened soil using
jet grouting, deep mixing or grouting techniques, the loading speed is reduced to
0.05 MPa/s for samples with an expected compressive strength fcyl,m ≤ 4 MPa.
This loading speed would also be in line with Plastic Concrete requirements and
achieve failure after approximately 20 s.
In Table 3.2 an overview of various testing methods for compressive strength
determination and their corresponding loading speed is given.
Table 3.2: Compressive strength testing methods and corresponding loading speed
Standard Type loading speed

DIN 18136 [62] strain rate (0.01·h0 )/min
ASTM D2166/D2166M [14] strain rate (0.005·h0 )/min to (0.02·h0 )/min
DIN EN 12390-3 [44] stress rate 0.6 ± 0.2 MPa/s
ASTM C39/39M [9] stress rate 0.25 ± 0.05 MPa/s
DIN 4093 [65] stress rate 0.05 MPa/s
h0 = initial sample height
3.2 Plastic Concrete Strength

As mentioned in section 3.1 the mechanical behaviour of concrete samples is
most commonly related to the samples’ compressive strength. However for cut-
off wall design the knowledge of Plastic Concrete’s tensile strength as well as
27
multi-axial strength is also of utmost importance. Therefore, in the following sub-

sections Plastic Concrete’s compressive, tensile and multi-axial strength will be
discussed and placed within the concrete technology context.
3.2.1 Unconfined Compressive Strength (UCS)

General
The strength of Plastic Concrete can be characterised with various parameters,

most commonly however the unconfined compressive strength (UCS) is here-
fore used. In concrete technology the w/c-ratio is the most common parameter
affecting concrete strength, whereby a lower w/c-ratio incurs in higher concrete
strength [79, 105]. Various studies have tested the UCS of Plastic Concrete with
varying mix designs [2, 22, 28, 72, 80, 93, 101, 110, 116]. In Figure 3.1 an overview
of the determined UCS from the aforementioned references is given. The ex-
perimental data plotted in Figure 3.1 corresponds to cylindrical Plastic Concrete
specimens with a height-to-diameter ratio of 2.0 (with varying size) produced with
common mixture compositions as described in section 2.2. The data shape indi-
cates which testing standard was used as can be seen in the graph legend.
6 ,0
u n c o n f in e d c o m p r e s s iv e s t r e n g t h ( M P a )
A lv a r e z 1 9 8 2
B a g h e r i2 0 0 8
5 ,0 B e c k e r2 0 1 5
E v a n s 1 9 8 7
H in c h b e r g e r 2 0 1 0
4 ,0 K a z e m ia n 2 0 1 6
M a h b o u b i2 0 0 5
P a s h a n g 2 0 1 2
3 ,0 S a d r e k a r im i2 0 0 2
e x p o n e n tia l f it
D IN E N 1 2 3 9 0 -3
2 ,0 A S T M C 3 9
D IN 1 8 1 3 6
A S T M D 2 1 6 6
1 ,0 o th e r / u n k n .
0 ,0
0 ,0 2 ,0 4 ,0 6 ,0 8 ,0 1 0 ,0 1 2 ,0 1 4 ,0 1 6 ,0
w /c - r a tio ( - )
Figure 3.1: Overview of the UCS of Plastic Concrete as a function of w/c-ratio at 28 days
The graph shows that, as would be expected, there is a gradual decline in

Plastic Concrete strength with increasing w/c-ratio, closely describing an expo-
nential trend. In addition some authors [72, 116] use a very high w/c-ratio far
exceeding commonly used w/c-ratios. However, due to the presence of ben-
tonite the effective w/c-ratio is smaller, since the bentonite absorbs water into
28
its structure reducing the readily available water for cement hydration. For this
reason Geil defines a reduced water cement ratio wred /c to account for the water
binding capacity of bentonite [76, p.45]. The author does not herefore provide
a mathematical formula, but instead provides a nomograph with which w/c-ratio
can be reduced as a function of cement content, bentonite content and bentonite
type [76, p.46]. It should here by again noted, that the water binding capacity of
bentonite is different for Na-based and Ca-based bentonite, as described in sec-
tion 2.1. The author however fails to analyse the contending behaviour of cement
and bentonite for the available water and the likely interaction between these.
If water adsorption capacity of bentonite is (simplistically) assumed to be 2.0
(in-line with section 2.1) the reduced water content wred can be estimated to
wred = w − 2 · mbentonite . With this, the overview in Figure 3.2 can be obtained.
6 ,0
u n c o n f in e d c o m p r e s s iv e s t r e n g t h ( M P a )
A lv a r e z 1 9 8 2
B a g h e r i2 0 0 8
5 ,0 B e c k e r2 0 1 5
E v a n s 1 9 8 7
H in c h b e r g e r 2 0 1 0
4 ,0 K a z e m ia n 2 0 1 6
M a h b o u b i2 0 0 5
P a s h a n g 2 0 1 2
3 ,0 S a d r e k a r im i2 0 0 2
D IN E N 1 2 3 9 0 -3
2 ,0 A S T M C 3 9
D IN 1 8 1 3 6
A S T M D 2 1 6 6
1 ,0 o th e r / u n k n .
0 ,0
0 ,0 1 ,0 2 ,0 3 ,0 4 ,0 5 ,0 6 ,0
w re d /c - r a tio ( - )
Figure 3.2: Overview of the UCS of Plastic Concrete as a function of wred /c at 28 days
It can be seen, that the wred /c relates to more realistic values ranging from
0.5 to 6.0. The expected exponential trend is no longer so clear for the data
as a whole, however a slight decrease in unconfined compressive strength with
decreasing wred /c-ratio can be recognised when looking into single data sets. It
does however stand out that, with some exceptions, the results obtained from
geotechnical testing standards ASTM D2166/D2166M [14] and DIN 18136 [62]
tend to be higher than those obtained from other testing standards, which implies
a testing induced difference. It furthermore becomes clear that the aforemen-
tioned simplistic approach to wred does not account for all influencing parameters
(e.g. bentonite type, bentonite characteristics, cement type, etc.). Moreover the
simplistic approach does not account for the effect of bentonite water adsorption
and the contending behaviour of cement and bentonite, which should therefore
be further analysed in more detail.
29
In concrete technology and design, standard concrete normally achieves a

fracture strain of approximately 0.2% to 0.3% when tested under standardised
unconfined compression conditions [79, p.378]. This guide value is however also
dependant on loading speed, whereby a slower loading speed further increases
the strain at failure obtained [79, p.379]. It is furthermore of common knowledge
that the fracture strain increases with increasing concrete strength, however the
post-crack behaviour is far more brittle the higher the concrete strength is [79,
p.379]. Plastic Concrete is therefore expected to have a higher fracture strain than
ordinary concrete and a far more ductile post-peak behaviour. This behaviour
has been corroborated by various studies, which identify an achievable strain at
failure for Plastic Concrete between 0.5% and 1.0% in unconfined compression
tests [2, 28, 99].
Strength Development
Although most reference testing is carried out at 28 days it is of common knowl-

edge in concrete science, that concrete strength continues to increase after 28
days. Concrete curing hereby mainly depends on the cement strength class, ce-
ment type and w/c-ratio used [79, p.325f.]. Blastfurnace cement (e.g. CEM III) de-
velops initial strength far slower than ordinary Portland cement (e.g. CEM I), how-
ever increases steadily far beyond the 28 day mark [105, p.67f.] [79, p.326f.]. This
is due to the latent hydraulic properties of blast furnace slag, which causes a slow
but steady strength development. The w/c-ratio also affects hydration rate of con-
crete, whereby with increasing w/c-ratio the hydration rate decreases [79, p.326].
Furthermore, the cement strength class also influences concrete strength devel-
opment, with higher cement strength classes causing a more rapid strength devel-
opment [79, 105]. For strength development of standard concrete, the fib Model
Code 2010 [88, p.87] gives an approximation for the time function of the con-
crete strength development as a function of the cement strength class, shown in
Equation 3.4.
βcc (t) = exp(s · [1 − (28/t)0.5 ]) (3.4)
where βcc (t) : time-dependant strength development function (-)

s : coefficient for cement strength class (-)
t : concrete age (d)
In line with these considerations, it can therefore expected that Plastic Con-
crete has a very low hydration rate due to the use of blastfurnace cement, a
low cement strength class and a high w/c-ratio. The effect of the cement type
30
and strength has also been shown to be predominant against the bentonite type
used [76, p.144f]. Various studies have examined the long-term strength of Plas-
tic Concrete mixtures [2, 22, 28, 89, 110]. In Figure 3.3 an overview of some of
these test results is given.
1 ,6
3 2 ,5 N
1 ,4
4 2 ,5 R
1 ,2
fcu(t) / fcu(2 8 d )
1 ,0
A lv a r e z 1 9 8 2 , f c u ,2 8
= 0 ,4 4 M P a , w /c = 3 ,8
0 ,8 B a g h e r i2 0 0 8 , f = 2 ,2 5 M P a , w /c = 1 ,8
c u ,2 8
t = 2 8 d
B e c k e r2 0 1 5 , f c u ,2 8
= 4 ,8 0 M P a , w /c = 2 ,2
0 ,6 P a s h a n g 2 0 1 2 , f c u ,2 8
= 2 ,6 8 M P a , w /c = 1 ,8
f ib M C - 2 0 1 0 , s = 0 , 2 0
f ib M C - 2 0 1 0 , s = 0 , 3 8
0 ,4
4 8 1 6 3 2 6 4 1 2 8 2 5 6 5 1 2 1 0 2 4
s p e c im e n a g e t ( d )
Figure 3.3: Overview of the UCS development as a function of time
As can be seen, the strength development of Plastic Concrete is not finalised

after 28 days, instead increasing steadily after 28 days. The studies also show
that due to the high w/c-ratio used the strength development of Plastic Concrete,
at any given cement strength class, is slower than the fib Model Code 2010 esti-
mates. It is also apparent that Plastic Concrete strength increases far beyond the
28 day mark and increases slowly before this date, which strongly relates to the
high w/c-ratio used. Alvarez et al.’s tests on Plastic Concrete (with a w/c-ratio of
3.78) for the Convento Viejo Dam project show that even after a 950 day testing
(not shown in Figure 3.3), the unconfined compressive strength continues to rise
and has not yet reached a plateau [2]. The compressive strength at 950 days
is hereby shown to be 5.59 times that at 28 days [2]. Furthermore, studies on
cement-treated soils and clays have also shown that the strength increase oc-
curs in a similar manner, with the strength increase being dependant on cement
type and cement content used [67, 127]. However, from the literature review, it
remains unclear how Plastic Concrete strength development affects the strain at
failure of samples, since contradictory results can be found. Some authors pro-
vide evidence that with increasing concrete strength (i.e. with sample age) the
strain at failure increases [28, 89]. Other authors however ascertain that with in-
creasing concrete strength the failure strain decreases [2, 99, 101]. Against the
31
background of concrete technology it should however be expected, that strain at

failure increases with increasing Plastic Concrete strength [79, p.379f.].
The knowledge of the long term strength development of Plastic Concrete is
of utmost importance, since cut-off walls are constructed for design periods far
exceeding 25 years. It is therefore not essential to test characteristic compressive
strengths of Plastic Concrete samples at 28 days and can instead be tested at
a higher age. Accordingly, the German standard DIN EN 1538 [47] therefore
states, that the knowledge of long-term strength and long-term deformability can
be necessary. In addition, the Austrian standard ÖNORM B4452 [108] requires
UCS testing to be performed with a sample age of less than 90 days, hereby
accepting testing beyond 28 days.
Caution is hereby advised, since a very low strength development may also
compromise the construction operation efficiency due to the alternating sequence
of primary and secondary panel construction and should therefore be considered
during the design phase.
Practical Application
The aforementioned testing was mainly carried out on purposely produced Plastic
Concrete specimens under laboratory conditions. In Europe, the concrete sam-
ples remain within the formwork for at least 16 h at an ambient temperature of
20 ± 5 °C. Once stripped, concrete samples are placed under water at controlled
temperature of 20 ± 2 °C until testing [42]. In Germany, the National Annex
requires concrete samples to cure under water for the 6 days subsequent to form-
work stripping. After this the samples may be stored in a climate room with tem-
peratures between 15 °C and 22 °C (ideally 20 ± 2 °C) and a relative air humidity
of 65 ± 5 % [43]. It is important to take these laboratory conditions into ac-
count when evaluating data, since laboratory specimens are cured with negligible
confining pressure and have an inexhaustible water supply during curing. On the
contrary, tremie-placed Plastic Concrete may encounter differing temperature and
humidity conditions. Some authors state for example that a curing temperature
of 10 °C or 15 °C is closer to the realistically encountered temperature within the
dam body [95, p.239] [55, p.17]. In addition, due to the reservoir water level Plas-
tic Concrete cut-off walls are constantly exposed to water, at least one-sidedly,
and should therefore be stored under water until required for testing whenever
possible [86, p.N36]. Furthermore, in-situ concrete is subjected to a confining
pressure applied by the adjacent ground and the overlying concrete which affects
the curing and consolidation of concrete specimens [80, 86].
Additionally, Plastic Concrete cut-off wall integrity is also affected by the tremie
method placement, since concrete may displace and destroy part of the filter cake
32
at the ground-bentonite interface in turn causing bentonite or soil inclusion into a

single concrete panel as well as water loss. The soil inclusions or water loss
may also vary depending on the surrounding soil, since different types of soil
have varying stiffness, water content or water permeability [5, 119]. This effect
is most commonly noted with cement-treated soils which are produced e.g. with
the cutter-soil-mixing (CSM) method, since water loss into the surroundings and
water content of soil cause local variations in w/c-ratio [5, 119].
3.2.2 Tensile Strength

Next to the unconfined compressive strength the uniaxial tensile strength (fct )
is an important parameter for the design of concrete structures. For standard
concrete the uniaxial tensile strength averages 10% of unconfined compressive
strength fcu [105, p.312]. This fct /fcu -ratio is however not constant and depends
on various effects. The ratio decreases with increasing compressive strength
fcu [105, p.311]. It furthermore decreases with time, in line with the increase
of compressive strength [105, p.311]. Furthermore, the ratio is affected by the
type of aggregate, aggregate grading, as well as curing conditions. The fct /fcu -
ratio is hereby higher for wet-cured than air-cured samples. For example fib
Model Code 2010, suggests that the mean tensile strength can be estimated from
the characteristic compressive strength using Equation 3.5 for concrete grades
≤ C50 [88, p.77].
fctm = 0.3 · (fck )2/3 (3.5)
where fctm : mean tensile strength (MPa)

fck : characteristic compressive strength (MPa)
Following Equation 3.5 for a Plastic Concrete sample with a UCS of 2.0 MPa
a mean tensile strength (fctm ) of 0.48 MPa should be expected, suggesting a
fct /fcu -ratio of 0.24. This is in line with the aforesaid deliberations, but differs sig-
nificantly from the often erroneously implemented fct /fcu -ratio of 0.10 for standard
concrete.
It should be however noted that the mean tensile strength (fctm ) refers to uni-
axial conditions, whilst tensile strength testing of concrete specimens most com-
monly occurs with the splitting tensile strength (fct,sp ) test. The European stan-
dard DIN EN 1992-1-1 suggests that fct /fct,sp -ratio can be assumed to be 0.9 [49,
p.28]. Recent studies have however shown, that a constant ratio is wrongfully
assumed. Malárics’ extensive study on standard and high performance concrete
shows that fct /fct,sp -ratio is inversely proportional to concrete strength, with the
33
ratio increasing with decreasing concrete strength [102, p.129ff.], as can be seen
in Figure 3.4.
Figure 3.4: Overview of the fct /fct,sp -ratio of concrete over compressive strength fcm (in
German) [102]
The conversion formula presented by Malárics, although limited to strength

classes ≥ C20, would suggest a fct /fct,sp -ratio between 1.86 and 2.35 for a UCS
of 2.0 MPa depending on specimen shape and aggregate type [102, p.130]. This
would imply, that for a fcu = 2.0 MPa Plastic Concrete with a tensile strength
fct = 0.48 MPa (as estimated with Equation 3.5), a splitting tensile strength of
approximately fct,sp ≈ 1.0 MPa should be obtained.
The literature review shows that, to date, studies into the tensile strength of
Plastic Concrete are scarce. Solely the USACE REMR GT-15 report [91] also
tests the splitting tensile strength of concrete. In this study, the authors as-
sumed that the splitting tensile strength (fctm,sp ) averages 26% of unconfined
shear strength (s) independently of the samples shear strength [91, p.94f.]. With
fu = 2 · s and an estimated fct /fct,sp -ratio of 2.0 (following [102]) the fct /fcu -ratio
can be estimated to:
fctm,sp fct /2.0 fct

0.26 = = ⇔ = 0.26
s fu /2 fu
This result would be in-line with the results obtained from Equation 3.5, how-
ever caution is advised since testing was only limited to on one set of samples
with a constant cement content of 300 lb/yd3 (178.0 kg/m3 ) with ages ranging from
3 days to 14 days and does not account for varying bentonite contents and w/c-
ratios [91, p.93]. In addition, as mentioned before, the fct /fcu -ratio is not constant
over concrete strength, suggesting possible testing differences, especially since
34
unconfined shear strength was determined with a triaxial testing apparatus. The
exact tensile strength to compressive strength relationship fct /fcu remains how-
ever unclear and should therefore be an important part of further investigations.
3.2.3 Multi-Axial Load-Bearing Capacity

Structural concrete is often exposed to multi-axial loading conditions within a
structure and the corresponding multi-axial load-bearing capacity of concrete is
therefore also of high importance. It should however firstly be remembered that
concrete failure under a uni-axial compressive force occurs through the inherent
development of a transversal tensile stress and the exceedance of the concrete
tensile strength [114, p.305ff.]. The concrete specimen hereby fails through the
development of cracks parallel to the direction of main loading exhibiting a brittle
behaviour [105]. This lateral strain may however be hindered through the ap-
plication of a compressive force perpendicular to the direction of main loading,
hereby increasing the overall compressive load-bearing capacity of a concrete
specimen [114]. By the contrary, if a perpendicular tensile strength is applied,
the overall compressive load-bearing capacity decreases. Similarly therefore if
a triaxial compression is applied with high lateral stresses, the concrete load-
bearing capacity increases manifold [105]. This increase is also known to be
more pronounced the lower the UCS and the lower the moisture content of con-
crete is [79, p.338]. The failure however no longer occurs through the exceedance
of tensile strength but instead through crushing, incurring in a change in failure to-
wards ductile behaviour [105, p.295]. Depending on the stress relationship of the
three stresses present concrete failure occurs through the development of shear
bands and shear failure [79, p.338f.] [114, p.306]. An overview of the failure mode
change depending on the stress applied can be seen in Figure 3.5.
The ductile failure hereby occurs at higher strain levels without the presence
of a strain-softening behaviour. This change in concrete failure is e.g. reported
by Sfer et al. [120], whereby at low confining pressure the failure occurs with
propagation of several distributed vertical and inclined cracks. On the contrary, at
higher confining pressure the response does not exhibit a well defined peak but
a monotonically decreasing slope tending towards a plateau, with failure occuring
through sudden propagation of the cracks [120].
In cut-off walls the Plastic Concrete is intrinsically submitted to a multi-axial
stress state. However, depending on dam settling and the upstream water level
the stress relationship may vary [82]. It is therefore of utmost importance to also
understand the multi-axial behaviour of Plastic Concrete. For Plastic Concrete a
similar behaviour to standard concrete can be expected. Since the uniaxial com-
35
Figure 3.5: Change in concrete failure mode depending on stress level relationship (in
German) [121]
pressive strength is low, the multi-axial load-bearing capacity increase can be

expected to be more pronounced. However, this increase is likely limited due to
the high water and moisture content of Plastic Concrete samples. Various studies
have also confirmed the change in failure mode with increasing confining pres-
sure for Plastic Concrete samples [28, 80, 89, 101]. At low confining pressures σc
cracking occurs parallel to specimen load axis [80], which some authors ascribe
to the progressive deterioration of cohesive bonds [101]. At higher confining pres-
sures specimen failure occurs through the existence of failure plane or a mixed
failure mode [80], which suggests the dominating frictional properties in speci-
men failure [28, 101, 110]. The specimens tested at higher confining pressures
not only exhibit a higher compressive load-bearing capacity and elastic modu-
lus [89], but also a more ductile, and possibly strain-hardening behaviour and an
overall higher strain at failure [28, 89, 101, 110]. An example of this change in
behaviour with increasing confining pressure can be seen in Figure 3.6.
Figure 3.6 shows that with increasing confining pressure, both the failure strain
and Plastic Concrete strength increase. Various studies have shown, that with
confining pressures between 200 kPa and 800 kPa a strain at failure between 2%
and 10% can be achieved, emphasising the highly ductile behaviour of Plastic
Concrete [28, 89, 101, 110].
Various authors, especially those from the geotechnical area, suggest that
Plastic Concrete may be considered a cohesive-frictional material for which a
Coulomb-type behaviour can be admitted, similar to very hard soils [101, 110].
Mohr-Coulomb failure criterion states that a material fails if a given combination of
normal and shear stresses exceeds the shear strength the material [123]. Accord-
36
Figure 3.6: Variation of deviator stress versus axial strain for unconfined and triaxial com-
pression tests [110]
ing to the principles of mechanics, the rupture line of the Mohr-Coulomb failure
criterion in a three-dimensional environment can therefore be described following
Equation 3.6 [117, p.151].
σ1 − σ3 σ1 + σ3
= c · cos ϕ + · sin ϕ (3.6)
2 2
where σ1 : greatest principle stress (MPa)
σ3 : smallest principle stress (MPa)
c : apparent cohesion (MPa)
ϕ : friction angle (°)
If no radial pressure σ3 is applied, the Mohr-Coulomb failure criterion given in

Equation 3.6 can be simplified to Equation 3.7.
2 · cos ϕ
σ1 = c · (3.7)
1 − sin ϕ
where σ1 : greatest principle stress (MPa)
c : apparent cohesion (MPa)
ϕ : friction angle (°)
Various studies have however shown that frictional angle ϕ and apparent co-
hesion c are not constant for any given Plastic Concrete mix. Hereby cohesion
parameter increases and frictional angle decreases with growing specimen age,
which the authors ascribe to cement hydration [101, 110]. An increase in cement
content further increases the cohesion parameter and decreases the friction an-
gle measured [101, 110]. Inversely, an increase in bentonite content decreases
the cohesion parameter and increases the frictional angle of Plastic Concrete
37
samples [110]. Some authors also suggest that a variation in the coarse-to-
fine aggregate ratio may change the measured parameters, with an increase in
coarse components increasing both apparent cohesion and friction angle [110].
An overview of the effects on mix-design variation on Mohr-Coulomb parameters
can be found in Table 3.3.
Table 3.3: Effect of mix-design changes on Mohr-Coulomb parameters
Increase in: cohesion c friction angle ϕ

specimen age + -
coarse-to-fine-ratio + +
bentonite content - +
cement factor + -
3.3 Elastic Modulus

The elastic modulus E of concrete is primarily determined by the elastic moduli
of its components cement paste and aggregate, as well as the volumetric propor-
tions of the materials in the mix, and may be estimated through composite the-
ory [114, p.310]. The elastic modulus of cement paste mainly depends on capillar
porosity and herewith on the w/c-ratio and degree of hydration. Aggregates gen-
erally have a higher elastic modulus which mainly depends on the mineralogical
properties of the rock. Therefore, generally speaking, an increase in water con-
tent or a decrease in cement content causes the elastic modulus of the obtained
concrete to decrease [114, p.310]. It is furthermore common knowledge that with
increasing degree of hydration the elastic modulus increases, whereby the elastic
modulus increase precedes the compressive strength increase [79, p.298]. This
is recognised in the fib Model Code 2010 since the time function of the elastic
modulus development of concrete βE (t) is estimated to be the square root of con-
crete strength development βcc (t) (see Equation 3.4), as shown in Equation 3.8.
βE (t) = [βcc (t)]0.5 (3.8)
where βE (t) : elastic modulus development function (-)

βcc (t) : strength development function (-)
Plastic Concrete behaves similarly to ordinary concrete. The elastic modulus

of Plastic Concrete increases with age and decreases with increasing bentonite
content [101,110]. In addition, the elastic modulus decreases with increasing w/c-
ratio, in-line with standard concrete behaviour [80, 101, 116, 135]. Some studies
38
have also shown, that the bentonite content also affects the elastic modulus devel-
opment over time, similarly to the strength development [101, 110]. Furthermore,
the elastic modulus further increases with increasing confining pressure during
testing in the triaxial testing apparatus [80, 101, 110]. In Figure 3.7 an overview of
the test results from various studies is given.
1 8 0 0
A lv a r e z 1 9 8 2
B A U E R
1 5 0 0 B e c k e r2 0 1 5
H in c h b e r g e r 2 0 1 0
e la s t ic m o d u lu s ( M P a )
K a z e m ia n 2 0 1 6
1 2 0 0 M a h b o u b i2 0 0 5
S a d r e k a r im i2 0 0 2
9 0 0 Z h a n g 2 0 1 3
D IN E N 1 2 3 9 0 -3
6 0 0 A S T M C 3 9
D IN 1 8 1 3 6
3 0 0 A S T M D 2 1 6 6
o th e r / u n k n .
0
0 ,0 2 ,0 4 ,0 6 ,0 8 ,0 1 0 ,0 1 2 ,0 1 4 ,0 1 6 ,0
w /c - r a tio ( - )
Figure 3.7: Overview of the elastic modulus as a function of the w/c-ratio at 28 days
As can be seen, the elastic modulus generally decreases with increasing w/c-
ratio, especially within one data series. However, at one given w/c-ratio the elas-
tic modulus scatters greatly. Taking into account that Zhang et al. [135] tested
the specimens following the Chinese concrete standard DL/T5150-2001 [122] it
may be suggested that the scatter may be induced by the testing method used.
In Figure 3.8 the elastic modulus is plotted over the corresponding compressive
strength.
Firstly, the elastic modulus increases with increasing compressive strength.

However, it hereby becomes apparent that the testing procedure used clearly in-
fluences the obtained elastic modulus, in-line with varying definitions of elastic
modulus given in section 3.1. The "elastic modulus" determined with concrete
testing standards (Zhang et al. [135]) is higher than that obtained from geotech-
nical testing standards (e.g. Mahboubi et al. [101]). This is most likely due to the
deformation measurement techniques used, since concrete standards measure
specimen deformation in-situ (e.g. strain gauges) while geotechnical standards
generally use the machine displacement to obtain specimen deformation. Simi-
lar results have been found in studies on cement treated soils [104, p.68f.]. This
39
1 8 0 0
A lv a r e z 1 9 8 2
B A U E R
1 5 0 0 B e c k e r2 0 1 5
H in c h b e r g e r 2 0 1 0
K a z e m ia n 2 0 1 6
1 2 0 0 M a h b o u b i2 0 0 5
S a d r e k a r im i2 0 0 2
Z h a n g 2 0 1 3
9 0 0
D IN E N 1 2 3 9 0 -3
A S T M C 3 9
6 0 0 D IN 1 8 1 3 6
A S T M D 2 1 6 6
o th e r / u n k n .
3 0 0
0
0 ,0 1 ,0 2 ,0 3 ,0 4 ,0 5 ,0 6 ,0
u n c o n f in e d c o m p r e s s iv e s tr e n g th ( M P a )
Figure 3.8: Elastic modulus as a function of the compressive strength at 28 days
further substantiates the fact, that the testing conditions should therefore be spec-
ified during planning and tendering of projects.
The fib Model Code 2010 for example suggests that the elastic modulus of
concrete may be estimated with Equation 3.9 [88, p.81].
1
Eci = Ec0 · αE · (fcm /10) 3 (3.9)
where Eci : modulus of elasticity at 28 days (MPa)

3
Ec0 : 21.5 · 10 (MPa)
αE : coefficient relating to aggregate type (-)
fcm : mean compressive strength at 28 days (MPa)
With this, a Plastic Concrete sample with a compressive strength of 2 MPa at

28 days could be estimated to have (with αE = 1.0) an elastic modulus of approx-
imately Eci ≈ 12573 MPa. It is however clear that, with decreasing compressive
strength the elastic modulus decreases disproportionally and therefore can not
simply be estimated following Equation 3.9. Based on the literature review, and
as shown in Figure 3.7 and Figure 3.8, the elastic modulus of Plastic Concrete
can be assumed to be in the range of 300 to 1500 MPa dependant on the testing
standard used.
In some cases, when the elastic modulus is determined as the deformation
modulus within a triaxial cell, the variation of deformation modulus over confin-
ing pressure should be accounted for. It can be expected, that similarly to the
compressive load-bearing capacity increase under confining pressure (see sub-
section 3.2.3), the elastic modulus will also increase with increasing confining
40
pressure. This is in-line with the results shown in Figure 3.9. The results also
show that, independently of the sample compressive strength, the elastic modu-
lus increases similarly. However, it should also be noted that some studies also
show a decreasing elastic modulus with increasing confining pressure. This phe-
nomena can however only be ascribed to an erroneous measurement of Plastic
Concrete specimen deformation.
1 0 0 0
8 0 0
H in c h .2 0 1 0 , fc u ,2 8
= 0 ,6 M P a
H in c h .2 0 1 0 , fc u ,2 8
= 0 ,9 5 M P a
6 0 0 H in c h .2 0 1 0 , fc = 2 ,5 M P a
u ,2 8
M a h b .2 0 0 5 , f c u ,2 8 = 3 ,0 2 M P a
P a s h .2 0 1 2 , f c u ,2 8 = 2 ,8 6 M P a
4 0 0
P a s h .2 0 1 2 , f c u ,2 8 = 3 ,6 6 M P a
B A U E R , f c u ,2 8 = 1 ,3 3 M P a
2 0 0
0
0 2 0 0 4 0 0 6 0 0 8 0 0 1 0 0 0
c o n f i n i n g p r e s s u r e σc ( k P a )
Figure 3.9: Elastic modulus as a function of confining pressure at 28 days
Some studies also investigate how the changes in Plastic Concrete compo-
sition may alter the resulting compressive strength [22, 36]. Bagheri et al. for
example studies the effect of cement substitution by silica fume [22]. It is es-
pecially interesting that with a 15% substitution of cement the Elastic Modulus
increases 180% for Plastic Concrete (w/c=1.8), while normal concrete (w/c=0.4)
would only increases 30% with the same substitution. As silica fume is known to
especially enhance the transition zone quality [105, p.668], the aforementioned
180% increase might suggest, that the transition zone is especially weak in Plas-
tic Concrete mixtures. The authors also state, that the for Plastic Concrete the
used of silica fume does not significantly alter the relationship between elastic
modulus and compressive strength.
As has been shown, the elastic modulus of concrete directly relates to com-
pressive strength. It is therefore important to note that the requirement of a
characteristic compressive strength fck (defined statistically as the 5-percentile
value) is not expedient since this automatically relates to an increase in the elas-
tic modulus [95, p.238]. It is therefore purposive to define a mean compressive
strength fcm which is required for the proposed cut-off wall and hereby also es-
tablish the targeted elastic modulus.
The literature review has also not presented any results regarding the Pois-
son’s ratio νc of Plastic Concrete. For ordinary concrete the Poisson’s ratio gen-
41
erally ranges between 0.15 and 0.25, is however mainly dependant on the stress
level reaching 0.5 at failure [114, p.311]. Within practical concrete stress levels,
the Poisson’s ratio of concrete is generally estimated to νc = 0.20 [88, 82]. How-
ever, due to the high water content in Plastic Concrete samples as well as the
presence of clay minerals such as bentonite causing high ductility of Plastic Con-
crete, the Poisson’s ratio is likely to be higher for such samples. As long as Plastic
Concrete is still considered an isotropic material, the shear modulus G should still
be estimated with Equation 3.10.
E
G= (3.10)
2 · (1 + ν)
where G : shear modulus of concrete (MPa)

E : elastic modulus of concrete (MPa)
ν : Poisson’s ratio of concrete (-)
Some authors have furthermore related the shear modulus G measured with
bender elements to the shear strength s or compressive strength of cement-
treated soils [119, 127]. Seng et al. for example suggest a near linear relation
G = 310 · s1.06 between shear modulus and shear strength [119, p.783]. With a
Poisson’s ratio ν = 0.20 and fcu = 2 · s [119, p.783] the relation E ≈ 350 · fcu
can be obtained, which is in-line with results shown in Figure 3.8. However, a
perfectly linear relationship is unlikely against the background of concrete tech-
nology (as shown in Equation 3.9) and further studies into this relationship should
be conducted.
3.4 Creep and Relaxation

When concrete is subjected to a load, concrete firstly reacts elastically. However,
besides elastic strain components, concrete also presents a non-linear stress-
strain behaviour. When subjected to sustained loading, strain increases gradually
with time due to concrete creep. The creep coefficient ϕ is hereby the most com-
mon engineering approach to estimate concrete creep and is defined following
Equation 3.11 [88, p.89] [114, p.320].
Various parameters affect the creep behaviour of concrete specimens. With
an increasing cement content and increasing water content, concrete creep in-
creases as it is the cement paste phase which undergoes creep [105, p.453].
Normal weight aggregates are generally not sensitive to creep, instead restrain-
ing concrete creep. This concrete creep restraint is more pronounced the higher
the elastic modulus of the aggregate is [114, p.322]. Some authors also suggest
42
that aggregate grading, maximum size and shape also affect concrete creep [105,
p.453]. Furthermore, creep is influenced by the ambient relative humidity, with
creep being higher, the lower the surrounding relative humidity [105, p.458] [114,
p.322]. In addition, concrete creep increases proportionally to stress within the
range of service stresses (normally σc < 0.4 · fcm ) [105, p.455] [114, p.320].
Last but not least, concrete creep is also dependant on the age at loading, with
creep increasing disproportionally the younger the concrete is at loading [4, p.23].
Therefore, depending on the conditions present the final creep coefficient ϕ∞
may vary greatly, normally ranging between 1 < ϕ∞ < 4 for standard con-
crete [114, p.321].
εcc (t, t0 )
ϕ(t, t0 ) = (3.11)
εci (t0 )
where ϕ(t, t0 ) : creep coefficient (-)

εcc (t, t0 ) : concrete creep strain (-)
εci (t0 ) : concrete elastic strain (-)
t0 : concrete age at loading (d)
Furthermore, independently of concrete loading, concrete water loss to its

surroundings causes concrete to dry out, in turn causing concrete specimens
to shrink. Since the magnitudes of shrinkage and creep are of the same order or
greater than those of elastic strain, these must be taken into account in the design
process [105, p.413ff.].
On the contrary, if a stressed concrete specimen is subjected to a constant
strain, the specimen stress will gradually decrease with time, known as relax-
ation. Similarly to the creep coefficient, the relaxation coefficient can be used to
measure the decrease in specimen stress following Equation 3.12.
∆σ(t, t0 )
ψ(t, t0 ) = (3.12)
σ0
where ψ(t, t0 ) : relaxation coefficient (-)
∆σ(t, t0 ) : stress decrease (MPa)
σ0 : initial stress (MPa)
Both creep and relaxation are based on the same molecular mechanisms and
therefore all influences affecting concrete creep also affect concrete relaxation.
It is herewith to possible to convert both coefficients into each other following
Equation 3.13 [79, p.416].
43
For a long duration of loading the relaxation parameter is set to ρ = 0.8 [114,
p.322]. It however remains to be proven, whether Equation 3.13 is also valid for
Plastic Concrete.
ϕ(t, t0 )
ψ(t, t0 ) = (3.13)
1 + ρ · ϕ(t, t0 )
where ψ(t, t0 ) : relaxation coefficient (-)

ϕ(t, t0 ) : creep coefficient (-)
ρ : relaxation parameter (-)
Taking into account the aforementioned influencing parameters, it should be

expected that Plastic Concrete has a greater creep and relaxation behaviour than
standard concrete. Various studies have confirmed these expectations [30, 80,
99]. Firstly, the very high w/c-ratio will likely incur in high water loss and speci-
men deformation. Furthermore, the relatively low elastic modulus of some of the
components in Plastic Concrete (e.g. bentonite) should further increase the creep
strain of Plastic Concrete specimens. In addition, due to the very slow strength
development of Plastic Concrete mixtures the specimen loading will likely occur
at a low degree of hydration furthering concrete creep. This behaviour is also
reported for similarly constituted cement-bentonite and cement-soil mixtures [55,
p.31]. Beckhaus et al. for example suggest a final creep coefficient ϕ∞ ≥ 2 for
Plastic Concrete samples, which they derive from results on soil samples solidi-
fied with the jet grouting technique [30, p.217f.]. It can be expected however that
Plastic Concrete mixtures may have even higher creep coefficients (e.g. ϕ∞ > 3).
Since concrete shrinkage uninterruptedly occurs, independently of load appli-
cation and water loss, the experimental set-up for relaxation testing has to con-
tinuously adjust the initially applied strain to compensate for concrete shrinkage.
Due to these demanding experimental requirements for relaxation testing, not
many studies exist relating to concrete relaxation [4, p.21].
Hinchberger et al. [80] studied the effect of constant axial strain on the stress
behaviour of Plastic Concrete and found that Plastic Concrete shows significant
stress relaxation effects with the measured stress reducing approximately 30%
after an 8 h period, as can be seen in Figure 3.10. The authors further note that
the stress relaxation process does not stabilise after the 8 h period, suggesting
that the long-term stress is likely to be even lower [80]. It should however be
noted, that Hinchberger et al. tested their specimens within a geotechnical, triax-
ial testing apparatus and did not compensate their results with the also inherent
concrete shrinkage, as mentioned before. On the other hand some studies have
shown that standard concrete has a strength relaxation of 20% after 8 hours [79]
44
as depcited in Figure 3.10. It is therefore necessary to conduct further testing in

this area to correctly establish the extent of creep and relaxation present in Plas-
tic Concrete samples. All in all, Plastic Concrete is expected to have a stronger
relaxation behaviour than standard concrete and therefore a time-dependant con-
stitutive model is required for Plastic Concrete [80].
1 ,1
H in c h b e r g e r 2 0 1 0 , fc u ,2 8
= 1 .0 M P a
H in c h b e r g e r 2 0 1 0 , fc u ,2 8
= 2 .0 M P a
G r ü b l2 0 0 1 , fc u ,2 8
= 4 1 .0 M P a
1 ,0
r e l . s t r e s s l e v e l σ( t ) /σ0
0 ,9
ε= 2 % = c o n s t.
0 ,8
0 ,7
0 ,0 1 0 ,1 1 1 0 1 0 0 1 0 0 0
tim e ( m in )
Figure 3.10: Stress relaxation of concrete samples over time
45
4. Hydraulic Behaviour
The seepage control of earth dams is the main purpose of a cut-off wall, as has
been illustrated in chapter 1. Hence, the hydraulic conductivity of Plastic Con-
crete is one of the most important parameters to be tested. Despite this, no spe-
cific testing standard exists for the measurement of Plastic Concrete permeability.
Therefore, standard test methods from geotechnical engineering as well as con-
crete technology are used, which are described in section 4.1. In section 4.2
various studies into the permeability of Plastic Concrete samples are presented.
4.1 Testing Methods

The hydraulic conductivity testing of concrete specimens can be foremost divided
into two main testing groups, namely those under loaded and unloaded condi-
tions. Hereby the measurement of hydraulic permeability occurs at simultaneous
load or after loading, respectively.
4.1.1 Unloaded Conditions

In concrete technology material hydraulic permeability is normally tested without
simultaneous loading. This can occur either on pristine, unloaded samples or on
previously loaded samples. The most common testing standard to determine the
water tightness of structural concrete samples is DIN EN 12390-8 [46], whereby
the depth of penetration of water under pressure is measured. In this, a concrete
specimen is placed within a pressure apparatus and submitted to a water pres-
sure 0.5 MPa for 72 h. The specimens are then split and the penetration depth
measured. Furthermore, in Germany, air-permeability of concrete is tested fol-
lowing DAfStb booklet 422 [34]. In this, the air permeability of concrete is tested
by pressing compressed air through a thin concrete disk in a one-dimensional
perfusion state [34].
Most interestingly Hoseini et al. reviewed and analysed many publications
regarding the effect of mechanical stress on permeability of concrete, whereby
the most of the data reviewed corresponds to testing methods where permeabil-
46
PLASTIC CONCRETE FOR CUT-OFF WALLS HYDRAULIC BEHAVIOUR
ity was measured after crack development [83]. The authors most importantly
noted that the permeability of concrete specimens depends on the applied stress,
whereby a threshold load level exists. With stresses below the threshold level the
permeability of concrete decreases due to the constriction of the pre-existing con-
duit network. However, above said threshold load level the permeability increases
swiftly due to the coalescence of micro-cracks. The authors however note, that
the threshold load level is not consistent amongst publications and is furthermore
not related to the compressive strength of concrete. Hoseini et al. [83] found that
permeability also depends on the crack geometry, whereby a threshold value for
crack width in the range of 50 µm to 100 µm exists. The authors also ascer-
tain that due to self-sealing in uncracked concrete and the autogenous healing
of cracks, the flow rate decreases with time [83, p.217]. They also state that the
concrete mix design affects the permeability of the concrete, as the aggregates
are far less permeable than the hydrated cement paste. The authors furthermore
note that permeability of concrete is far more dependant on the permeability of its
constitutients than on the cement-aggregate interface [83].
Hoseini et al. state that all the tests reviewed vary due to 1) lack of equilibrium
in the fluid flow and 2) most of the data is from permeability being measured after
cracking. Therefore, future research must account for the effect of thermal cycles
on mass transport, as fluid takes days to reach equilibrium. Finally research must
focus on a threshold crack width rather than a threshold stress level to determine
the onset of critical levels of fluid permeability [83].
4.1.2 Loaded Conditions

In various practical applications (including Plastic Concrete) concrete is submitted
to compressive or flexural forces while simultaneously being permeated through.
Despite this, the aforementioned testing methods are solely capable of measuring
concrete permeability in unloaded conditions. However, a few testing methods ex-
ist with which permeability can be measured during loaded conditions for concrete
and soil.
Most notably, geotechnical triaxial cells may be used for this purpose. In Ger-
many, the described method is standardised in DIN 18130-1 [58] for laboratory
conditions and DIN 18130-2 [59] for field conditions. A sample (most commonly
soil) is placed within the triaxial cell and subsequently compressed in axial di-
rection. The specimen is then permeated in axial direction with de-aired water
and the resulting flow volume is measured at set intervals. With the given testing
geometry, hydraulic gradient i and temperature, the hydraulic conductivity can be
determined [118].
47
Hydraulic conductivity is hereby defined using Darcy’s law, which describes

the flow of a fluid through a porous medium. This is most commonly used to
describe the permeability of soil, whereby an incompressible fluid with constant
volume is assumed [117, p.79] and laminar flow within fully saturated samples is
implied [118]. In this case, Darcy’s law simplifies to Equation 4.1.
v
kf = (4.1)
i
where k : permeability coefficient (m/s)
v : filtration velocity (m/s)
i : hydraulic gradient (-)
Normally the permeability of soil is determined in triaxial cells with water flow-
ing through sample from bottom to top. The samples have to be fully saturated
and therefore water transport through convection can be assumed [118]. Only the
capillary pores and the water contained herein are available for convective water
transport. The mineral compound can also be considered impermeable [118].
For standard concrete specimens some authors also showed that at low com-
pressive pressures a minor decrease in the measurement of specimen permeabil-
ity occurs [23]. However, as mentioned before, a threshold value of crack width
seems to exist, after which a significant increase in permeability can be expected.
In addition, the authors state that the loading history appears to be a critical fac-
tor controlling the permeation through stressed concrete. They also note that in
their study a decrease in the permeability coefficient occured with time, which the
authors ascribe to continued hydration as well as potential pore blocking [23].
It should be noted however that DIN 18130-1 [58] and DIN 18130-2 [59] are
both designed for testing granular soils and not for cut-off wall materials or con-
crete [118]. The main difference between soils and cut-off wall materials (COWM)
is their distribution of air voids, since COWM have a larger air void content, which
is however filled with water. Soil on the other hand has larger amounts of empty
air voids. DIN 18130 [58, 59] states that the de-airing of water should be con-
ducted and the specimen thoroughly saturated [118]. Scholz et al. however
showed that, unlike for soils, the deairing of water is not strictly necessary for
COWM since the result with non-deaired water are only minimally lower [118]. In
addition their study showed no difference between fully-saturated and partially-
saturated COWM specimens, which the authors ascribe to the material-intrinsic
water-saturation [118].
For concrete specimens other authors have also studied the effect of load
level on water permeability [134]. The authors hereby attempt to determine the
water permeability of a rectangular concrete specimen by applying flexural load
48
using a centre-placed fulcrum. It is thereby shown that with increasing flexural

load the water permeability increases [134]. This study also shows that when the
load level exceeds 30% of the crack load fcr the relative permeability increases
remarkably. This study should however be considered with caution, as Ø 8 mm
reinforcement bars were placed within the concrete specimens, likely affecting the
crack formation and also differing from non-reinforced Plastic Concrete behaviour.
4.2 Plastic Concrete Permeability

As mentioned previously, some few studies exist into the hydraulic conductivity of
Plastic Concrete Mixtures [2, 22, 28, 72, 80, 101]. It should hereby be noted that,
due its low strength, degree of water-tightness and composition, Plastic Concrete
is commonly tested following geotechnical testing standards and not structural
concrete penetrations tests. An overview of some test results of hydraulic con-
ductivity without confining pressure is given in Figure 4.1.
-3
1 0
A lv a re z 1 9 8 2
1 0 -4 B a g h e r i2 0 0 8
h y d r a u lic c o n d u c tiv tiy k ( m /s )
B e c k e r2 0 1 5
1 0 -5 E v a n s 1 9 8 7
te n d e n c y
-6 D IN 1 8 1 3 0 -1
1 0
o th e r / u n k .
-7
1 0
-8
1 0
-9
1 0
-1 0
1 0
-1 1
1 0
0 .0 1 .0 2 .0 3 .0 4 .0 5 .0 6 .0
w re d /c - r a tio ( - )
Figure 4.1: Hydraulic conductivity of Plastic Concrete over wred /c-ratio at σc = 0
It can be seen that, similarly to standard concrete, the hydraulic conductivity

of Plastic Concrete specimens increases with increasing wred /c-ratio. This may be
ascribed to a reduced particle-cross linking and an increased air void content with
increasing w/c-ratio. Similarly, when hydraulic conductivity of Plastic Concrete is
related to its unconfined compressive strength a decrease in permeability can be
observed with increasing compressive strength as shown in Figure 4.2.
The results should however be considered with caution due to the small basis
of data available. Plastic Concrete hydraulic conductivity hereby ranges between
10-10 m/s and 10-7 m/s. Similarly the hydraulic conductivity of Plastic Concrete
49
-5
1 0
B a g h e r i2 0 0 8
-6 B e c k e r2 0 1 5
1 0
h y d r a u lic c o n d u c tiv ity k ( m /s )
E v a n s 1 9 8 7
-7
C h e n g 2 0 1 2
1 0 lin e a r r e g r e s s io n
D IN 1 8 1 3 0 -1
-8
1 0 o th e r / u n k .
-9
1 0
-1 0
1 0
-1 1
1 0
-1 2
1 0
0 .0 2 .0 4 .0 6 .0 8 .0 1 0 .0
u n c o n fin e d c o m p r e s s iv e s tr e n g th ( M P a )
Figure 4.2: Hydraulic conductivity of Plastic Concrete over compressive strength
is consistently lower than that of cement-bentonite diaphragm wall materials [72].

Mahboubi et al. hereby suggest that increasing cement factor decreases perme-
ability, as more water is consumed during cement hydration and less free water
remains [101]. Some authors also note that an addition of fly-ash may further
reduce permeability, a partial cement substitution by fly-ash on the other hand
increases hydraulic conductivity [72]. However, to date, the effect of substitution
has only been studied at 14 days, for which reason more detailed and extensive
research is necessary. In another study the effect of silica fume on Plastic Con-
crete hydraulic conductivity was studied [22]. The results showed that through
the use of silica fume, and the increase in w/c-ratio to maintain the strength level,
a ten-fold reduction in permeability is possible [22]. However, the economic effi-
ciency should be always considered when implementing such substitutions.
Some authors also study the hydraulic conductivity of Plastic Concrete under
loaded conditions [80]. Most notably, Hinchberger et al. studied the effect of
axial strain on hydraulic conductivity of Plastic Concrete within a triaxial cell. In
Figure 4.3 an extract of their results is shown. The authors hereby tested various
Plastic Concrete mixtures with distinct compressive strengths at varying confining
pressures.
As can be seen in Figure 4.3, hydraulic conductivity increases with increas-
ing axial strain at a given compressive strength and confining pressure. It should
be noted that the low strength Plastic Concrete mixtures (fcu = 0.95 MPa) follows a
relatively smooth curve, whilst the higher strength Plastic Concrete (fcu = 2.5 MPa)
behaves differently. At low axial strain the stronger concrete, with a possibly more
dense structure, exhibits a lower hydraulic conductivity. However, with increasing
strain, the hydraulic conductivity increases significantly surpassing that of lower
50
-3
1 0
f c u = 0 . 9 5 M P a , σc = 0 . 1 0 M P a
h y d r a u lic c o n d u c tiv ity k ( m /s ) f c u = 2 . 5 0 M P a , σc = 0 . 1 0 M P a
-4
1 0 f c u = 0 . 9 5 M P a , σc = 0 . 9 0 M P a
f c u = 2 . 5 0 M P a , σc = 0 . 9 0 M P a
-5
1 0
-6
1 0
-7
1 0
-8
1 0
-9
1 0
0 .0 2 .0 4 .0 6 .0 8 .0 1 0 .0
a x ia l s tr a in ( % )
Figure 4.3: Hydraulic conductivity of Plastic Concrete with distinct compressive

strengths fcu at varying confining pressures σc (following [80])
strength Plastic Concretes at identical axial strains. Similarly to concrete [83],

these test results suggest the existence of a threshold strain or threshold stress,
whereby the surpassing of this threshold significantly increases hydraulic conduc-
tivity.
In addition, the influence of the confining pressure is clearly visible, whereby
at higher confining pressure the hydraulic conductivity decreases [80, figs.14-16],
which various authors relate to possible pore closing and the pressing together of
concrete crack boundaries [80,101]. In all cases however, the combined compres-
sion and permeation may cause side-wall leakage during testing which should be
accounted for and eliminated through the application of impermeable materials
on the lateral surface areas of Plastic Concrete specimens [80].
Furthermore it should be noted that current design procedure for Plastic Con-
crete does not account for the highly ductile behaviour of this material, whereby
a high relaxation and creep potential have been shown to exist (see section 3.4).
This behaviour is beneficial for Plastic Concrete hydraulic permeability, since
it can prevent material stress peaks during loading and avoid the formation of
cracks, which would incur in an increase in permeability. Some initial studies
have shown that with deformation of approximately 70% of strain at failure, no
significant increase in hydraulic conductivity occurs [99, p.101]. By contrast crack
onset in concrete generally occurs at approximately 20% of strain [99].
Finally, some authors suggest that the addition of bentonite reduces the hy-
draulic conductivity of Plastic Concrete samples [74, 76, 101, 118]. In Figure 4.4
an overview of the hydraulic conductivity over bentonite content is given.
It becomes apparent that the hydraulic conductivity of Plastic Concrete may
51
-3
1 0
A lv a re z 1 9 8 2
1 0 -4 B a g h e r i2 0 0 8
h y d r a u lic c o n d u c tiv ity k ( m /s )
B e c k e r2 0 1 5
1 0 -5 E v a n s 1 9 8 7
-6 D IN 1 8 1 3 0 -1
1 0
o th e r / u n k .
-7
1 0
-8
1 0
-9
1 0
-1 0
1 0
-1 1
1 0
0 .0 0 .5 1 .0 1 .5 2 .0 2 .5 3 .0 3 .5 4 .0 4 .5
b e n to n ite c o n te n t ( w t- % )
Figure 4.4: Hydraulic conductivity of Plastic Concrete over bentonite content at σc = 0
not be directly related to bentonite content, since the data shows no clear ten-
dency. Some authors further suggest that an optimum specific bentonite content
exists, at which the permeability reaches its minimum before increasing again [101].
This can however not be confirmed with the scarce amount of data available in lit-
erature and highlights the need for further, systematic research into the hydraulic
conductivity of Plastic Concrete.
Only few studies also refer to the time-development of Plastic Concrete hy-
draulic conductivity [28, 94]. The authors hereby ascertain that with increasing
time the hydraulic conductivity decreases [28, 94]. This is in line with the strength
development behaviour of Plastic Concrete (see subsection 3.2.1) and is likely
caused by the progress of hydration and the formation of CSH phases consoli-
dating cement particles together [84]. In addition, crack self-healing and crack ob-
struction with the transported particles, amongst others, are also known to cause
the permeability of concrete to further decrease over time [69, p.35].
However, some authors suggest that depending on Plastic Concrete mix de-
sign and materials used, the hydraulic conductivity may increase slightly at higher
age [28], which may be related to mechanical and chemical actions on Plas-
tic Concrete cut-off walls. Some specifications therefore allow for permeability
tests at higher ages (e.g. 90 days), to achieve the required design permeabil-
ity values [86, p.N13]. The Austrian standard OENORM B 4452 for example
allows Plastic Concrete permeability testing to be performed up to 90 days of
age [108, p.9]. It may however also be contractually expedient to set 28-day
control values, not as the design permeability but as a demonstration of design
value achievement, to shorten the acceptance period of the construction services
provided [86, p.N13].
52
5. Summary and Outlook
5.1 Summary
With the present report first steps are set out for a comprehensive understanding
of Plastic Concrete material behaviour. It can be concluded that Plastic Concrete
is a low strength concrete due to its material behaviour, mix design and place-
ment method. With the acquired knowledge Plastic Concrete can be used to
safely guarantee seepage control inside and below dams with a controlled mate-
rial behaviour. All in all, the following considerations may be taken into account
for Plastic Concrete cut-off wall design.
5.1.1 Mix Design

Plastic Concrete can be considered to be a low strength concrete with a low elastic
modulus capable of sustaining larger strains than normal concrete. These prop-
erties can be achieved through the targeted selection of source materials and mix
design. Equally to standard concrete, cement is used as a binder in Plastic Con-
crete. Although ordinary Portland cement is most commonly used, blast-furnace
or pozzolanic cement may also be used. These types of cement are known to
incur in slower concrete strength development and lower permeability of the hard-
ened material. The key component differentiating Plastic Concrete from ordinary
concrete is the far higher w/c-ratio, for which the fresh concrete stability has to be
controlled by low amounts of physically water-binding additions. Most commonly
bentonite, a clay-rock composed of montmorillonite minerals, is added since ben-
tonite is known to adsorb a great amount of water and hence is implemented as
a stabilising agent. To date, bentonite characterisation is however only limited to
performance-based testing procedures. Possible water-binding additions are not
however limited to bentonite. Finally, Plastic Concrete uses regular aggregate for
its mix design with the maximum aggregate size being often limited to 16 mm
due to the segregation risk with greater maximum grain size. As admixtures re-
tarding agents are commonly implemented to delay concrete setting in tremie
placement, whilst super-plasticizing admixtures are sometimes used to increase
concrete workability.
53
PLASTIC CONCRETE FOR CUT-OFF WALLS SUMMARY AND OUTLOOK
Plastic Concrete mix design is similar to that of standard concrete. The ag-
gregate content ranges from 1300 to 1900 kg/m3 . Cement content is normally in
the range of 80 to 200 kg/m3 . The w/c-ratio generally ranges between 2.0 and
5.0, however the exact value depends mostly on the target strength and source
materials used. The mixing sequence also affects material behaviour, whereby
currently no common mixing sequence exists. Usually, bentonite and water are
premixed to a bentonite slurry. The bentonite slurry is then allowed to hydrate for
a period of time, between 3 hours and 24 hours. After this, the bentonite slurry
is mixed with cement, sand and gravel. Alternative mixing procedures do how-
ever exist and there influence on the final material properties should be the aim
of future studies.
5.1.2 Mechanical Behaviour

The mechanical behaviour of Plastic Concrete is in line with that which can be
expected from concrete technology. It should however be noted that much testing
is conducted using geotechnical testing standards and not concrete testing stan-
dards. This difference is especially important when testing the elastic modulus of
Plastic Concrete to asses the materials’ deformability.
Generally speaking, it can be ascertained that the compressive strength of
Plastic Concrete increases with decreasing w/c-ratio. It should be noted however
that the w/c-ratio does not account for the addition of bentonite and therefore not
consider the reduction in free water available for cement hydration. Plastic Con-
crete compressive strength normally lies between 0.5 to 2.5 MPa at 28 days. In
addition to this, it is known that the compressive strength development is very
pronounced for Plastic Concrete, far beyond the 28 day mark. The magnitude of
strength development clearly depends on the cement type used (ordinary Port-
land, blast-furnace or pozzolan cement). It may therefore also be expedient to
test Plastic Concrete compressive strength at higher ages, e.g. 90 days. Fur-
thermore, for the purpose of simplification, it is also recommended to establish a
minimum compressive strength of at least 0.3 MPa at 28 days to ensure erosion
stability of Plastic Concrete mixtures.
The strain at failure of Plastic Concrete is also far greater than that of stan-
dard concrete, where under compression a maximum strain of up to 1% can be
achieved. It may therefore be expedient to establish a maximum strain at failure
in the range of 0.8% to 1.0% as a requirement. It should however be noted that
concrete behaves softer (i.e. has a lower Young’s modulus) the lower the com-
pressive strength is.
54
The tensile to compressive strength ratio of Plastic Concrete is also expected

to be greater than that of standard concrete, with the tensile strength being in the
range of 10% to 20% of compressive strength. Under multi-axial load, the load
bearing capacity clearly increases with axial strains as high as 10%.
The magnitude of the elastic modulus of Plastic Concrete clearly depends on
the testing standard used. On the one hand the definition of elastic modulus is dif-
ferent for concrete standards (i.e. defined as the secant modulus within a testing
procedure) and geotechnical standard (i.e. defined as the maximum tangential
modulus in unconfined compression testing). The deformation is also measured
inconsistently with the geotechnical standards commonly measuring piston move-
ment or occasionally measuring loading plate movement (ex-situ) whilst con-
crete standards require deformation measurement of the sample through strain
gauges, LVDTs or similar (in-situ). The deformation modulus (geotechnical stan-
dard) of Plastic Concrete can therefore be estimated to 100-600 MPa, whilst
Young’s modulus (concrete standard) should be estimated in the range of 300-
1800 MPa. In addition, as would be expected, the elastic modulus increases with
increasing confining pressure.
Since Plastic Concrete is generally mixed with a high w/c-ratio, the creep and
relaxation properties are more pronounced than those of standard concrete. With
this the final creep coefficient can be expected to be ϕ∞ ≥ 3.0. Therefore the
relaxation potential of Plastic Concrete is also notably higher than that of stan-
dard concrete, which is however not accounted for in Plastic Concrete design.
The higher relaxation potential of Plastic Concrete is in turn beneficial to prevent
material stress peaks during loading and avoid the formation of cracks, which
would incur in an increase in permeability. It should be underlined that, to date,
the relaxation behaviour is not taken into consideration in material design.
5.1.3 Hydraulic Behaviour

The hydraulic behaviour of Plastic Concrete, and concrete in general, remains a
relatively unstudied field, especially for testing under realistic stress conditions.
The testing of hydraulic conductivity is hereby not standardised for concrete,
whereby two groups of testing conditions can be differentiated, namely testing
under loaded and unloaded conditions. Most commonly testing is conducted
under unloaded conditions, whereby concrete specimens have sometimes been
preloaded prior to testing. Various studies have shown that concrete permeability
mainly depends on a threshold crack value, not a threshold stress value, after
which permeability increases significantly.
55
For Plastic Concrete it has been shown that permeability decreases with de-
creasing w/c-ratio which is linked to a less porous material structure. The litera-
ture review has also shown that an increase in bentonite quantity alone can not
be related to a decrease in Plastic Concrete permeability.
In a triaxial cell, permeability testing can be conducted with simultaneous tri-
axial compression. When testing under these conditions it can be seen that a
threshold value exists after which permeability of Plastic Concrete changes. High
strength Plastic Concrete (> 2.5 MPa) has a lower initial permeability however un-
der increased strain, permeability increases far more significantly than for lower
strength Plastic Concrete. These results should however be considered with cau-
tion, since permeability testing in triaxial cells is designed for soil testing and has
not been studied in detail for concrete specimens. Furthermore, as would be ex-
pected, an increase in confining pressure has shown to reduce Plastic Concrete
permeability.
The change in Plastic Concrete permeability over time is scarcely reported in
literature, however a decrease in permeability over time has been shown to exist.
This is in line with concrete technology where with progressive cement hydration
the permeability decreases through the increasing cross-linking of particles. This
is especially relevant with blast-furnace or pozzolan cements where increased
cross-linking occurs at higher ages (pozzolan effect). It is therefore expedient
that Plastic Concrete permeability testing is conducted at ages greater than 28
days (e.g. 90 days) to account for the permeability increase with time. This in
turn represents the long-term material behaviour of Plastic Concrete more realis-
tically. Plastic Concrete permeability can therefore be estimated in the range of
1 · 10−8 m/s to 1 · 10−9 m/s depending on testing age.
5.2 Future Research

With the present report the State-of-the-Art for Plastic Concrete in cut-off walls
has been established. Despite these promising results, questions remain which
should be the purpose of further studies.
Firstly, further research is required to examine the effects of the mixing proce-
dure on Plastic Concrete hardened behaviour. The focus of these studies should
be placed at understanding the interaction of water, bentonite and cement and
to what extent the varying mixing procedures may alter the availability of water
during cement hydration. Further studies into the mix design and the source
materials used in Plastic Concrete may also identify optimisation possibilities for
Plastic Concrete material behaviour. Most notably reliable analytical methods
must be studied to comprehensively characterise bentonite source materials as
56
this may shed light on the mechanism underlying Plastic Concrete behaviour and
establish bentonite requirements. The understanding of these mechanisms is
also of utmost importance to establish their influence on compressive and tensile
strength of Plastic Concrete as well as creep behaviour.
Furthermore, research is needed to determine to what extent bentonite af-
fects the final material behaviour and establish the effectiveness of bentonite to
increase Plastic Concrete ductility and deformability. In addition, the long-term
behaviour of Plastic Concrete should be further studied to determine permeability
and compressive strength change over time.
Of transcendent importance is the need to further investigate the creep and
relaxation potential of Plastic Concrete, since these have a significant impact on
the material stress. This in turn strongly affects cut-off wall design since a high
relaxation potential provides a far greater potential deformability before incuring in
cut-off wall damage. In addition, further research should be undertaken to confirm
the increased tensile to compressive strength ratio present for Plastic Concrete.
Finally, a comparative, experimental study between concrete and geotechnical
testing standards is of high interest to enable the transposition of test results into
one another and determine the causes for the disparate test results.
On the other hand, the permeability changes in Plastic Concrete should be the
subject of further studies. A greater focus on the determination of Plastic Con-
crete permeability under simultaneous loading could produce important findings
that account for a more realistic design of Plastic Concrete cut-off walls. For this,
the development of a new testing method may also be necessary. Further studies
are also need to be carried out to validate the reported decrease in Plastic Con-
crete permeability over time, whereby a distinction of the source materials used
should be made. It would also be interesting to assess the effects of different ben-
tonites on the Plastic Concrete permeability values and relate these to bentonite
structure characteristics. Last but not least, further studies into the identification
of threshold strain value for Plastic Concrete permeability increase must be con-
ducted to safely design cut-off walls and ensure their imperviousness to water
during operation.
All in all it may be summarised that the findings of this study have a num-
ber of important implications for future practice. However, continued efforts are
needed to further understand Plastic Concrete behaviour and ensure its correct
application in cut-off wall design.
57
Acknowledgements
The authors of this report would like to thank the BAUER Forschungsgemein-
schaft (BFG) for the financial support provided for this project.
58
Bibliography
[1] H. Alehossein, K. Beckhaus, and M. Larisch. Analysis of L-box test for

tremie pipe concrete. ACI Materials Journal, 109(3):303–312, 2012.
[2] L. Alvarez, J. Larenas, A. Bernal, and J. A. Marin. Characteristics of the

Plastic Concrete of the Diaphragm Wall of Convento Viejo Dam. In Interna-
tional Commission On Large Dams (Ed.), 14th International Congress on
Large Dams in Rio de Janeiro, Brazil, volume IV, pp. 371–389. 1982.
[3] K. Anastasopoulos, C. Economidis, D. A. Bruce, and G. Rizopoulos. Con-

struction of Papadia secant pile plastic concrete cut-off wall. In A. Anag-
nostopoulos, M. Pachakis, and C. Tsatsanifos, editors, Proceedings of the
15th European Conference on Soil Mechanics and Geotechnical Engineer-
ing, pp. 1105–1110. IOS Press, Amsterdam, 2011.
[4] I. Anders. Stoffgesetz zur Beschreibung des Kriech- und Relaxationsver-

haltens junger normal- und hochfester Betone: Dissertation, volume 73
of Karlsruher Reihe Massivbau, Baustofftechnologie, Materialprüfung. KIT
Scientific Publishing, Karlsruhe, 2013.
[5] M. Arnold, K. Beckhaus, and U. Wiedenmann. Cut-off wall construction

using Cutter Soil Mixing: A case study. geotechnik, 34(1):11–21, 2011.
[6] ASTM International. ASTM C1157/C1157M-11 - Performance Specification

for Hydraulic Cement.
[7] ASTM International. ASTM C150/C150M-16e1 - Specification for Portland

Cement.
[8] ASTM International. ASTM C219-14a - Terminology Relating to Hydraulic

Cement.
[9] ASTM International. ASTM C39/C39M-17 - Test Method for Compressive

Strength of Cylindrical Concrete Specimens.
[10] ASTM International. ASTM C469/C469M-14 - Test Method for Static Mod-
ulus of Elasticity and Poissons Ratio of Concrete in Compression.
[11] ASTM International. ASTM C595/595M-17 - Specification for Blended Hy-

draulic Cements.
[12] ASTM International. ASTM C617/C617M-15 - Practice for Capping Cylin-

drical Concrete Specimens.
59
PLASTIC CONCRETE FOR CUT-OFF WALLS BIBLIOGRAPHY
[13] ASTM International. ASTM C801-98 - Test Method for Determining the
Mechanical Properties of Hardened Concrete Under Triaxial Loads.
[14] ASTM International. ASTM D2166/D2166M-16 - Test Method for Uncon-

fined Compressive Strength of Cohesive Soil.
[15] ASTM International. ASTM D2435/2435M-11 -Test Methods for One-

Dimensional Consolidation Properties of Soils Using Incremental Loading.
[16] ASTM International. ASTM D2850-15 - Test Method for Unconsolidated-

Undrained Triaxial Compression Test on Cohesive Soils.
[17] ASTM International. ASTM D4318-10e1 - Test Methods for Liquid Limit,
Plastic Limit, and Plasticity Index of Soils.
[18] ASTM International. ASTM D4767-11 -Test Method for Consolidated

Undrained Triaxial Compression Test for Cohesive Soils.
[19] ASTM International. ASTM D5890-11 - Test Method for Swell Index of Clay
Mineral Component of Geosynthetic Clay Liners.
[20] ASTM International. ASTM D6910-09 - Test Method for Marsh Funnel Vis-
cosity of Clay Construction Slurries.
[21] ASTM International. ASTM D7181-11 - Test Method for Consolidated

Drained Triaxial Compression Test for Soils.
[22] A. R. Bagheri, M. Alibabaie, and M. Babaie. Reduction in the permeability

of plastic concrete for cut-off walls through utilization of silica fume. Con-
struction and Building Materials, 22(6):1247–1252, 2008.
[23] N. Banthia, A. Biparva, and S. Mindess. Permeability of concrete under

stress. Cement and Concrete Research, 35(9):1651–1655, 2005.
[24] P. Banzhaf. Erfahrungen bei der Herstellung der Zweiphasen-Dichtwand

Hinze-Damm, Australien. In C. Bergmann (Ed.), Vorträge zum 17. Darm-
städter Geotechnik-Kolloquium, Mitteilungen des Institutes und der Ver-
suchsanstalt für Geotechnik der Technischen Universität Darmstadt, pp.
39–48. Darmstadt, 2010.
[25] P. Banzhaf. Dichtwandherstellung für den Bagatelle-Damm auf Mauritius.

bbr Leitungsbau, Brunnenbau, Geothermie, 67(8):36–41, 2016.
[26] P. Banzhaf and A. Seidel. Die Bauverfahren zur Deichsanierung. Tiefbau,

16(10):626–627, 2004.
[27] BAUER Spezialtiefbau GmbH. BAUER Diaphragm and Cut-Off Walls-

Technical Brochure, 05/2015.
[28] A. Becker and C. Vrettos. Laboruntersuchungen zum Materialverhalten von

Tonbeton. Bautechnik, 92(2):152–160, 2015.
60
[29] K. Beckhaus. "Plastic Concrete" - Erfüllung schwieriger Anforderungen

an Baustoffe des Spezialtiefbaus. In N. Vogt (Ed.), Wasser und Boden,
Beiträge zum 10. Geotechnik-Tag in München, pp. 21–29. Technische Uni-
versität München, 2011.
[30] K. Beckhaus, H. Lesemann, P. Banzhaf, and C. Högl. Welches Stoffmodell

für hochverformbaren Dichtwandbeton ist geeignet für die geotechnische
Bemessung im Dammbau? In Österr. Ingenieur- und Architekten-Verein
(Ed.), Tagungsbeiträge zur 10. Österreichischen Geotechniktagung, pp.
211–222, Wien, 2015.
[31] K. Beckhaus, S. Schmitz, and W. Schwarz. Deep Cut-Off Walls constructed

under Dams with Trench Cutters. In Proceedings of the 2nd International
Conference on Long Term Behaviour of Dams - LTBD09, pp. 305–310,
Graz, 2009.
[32] C. Boley and D. Adam. Handbuch Geotechnik: Grundlagen - Anwendun-

gen - Praxiserfahrungen. Vieweg+Teubner Verlag, Wiesbaden, 1st Edition,
2012.
[33] D. A. Bruce (Ed.). Specialty construction techniques for dam and levee
remediation. CRC Press, Boca Raton, 2013.
[34] N. Bunke. Prüfung von Beton: Empfehlungen und Hinweise als Ergänzung
zu DIN 1048, volume 422 of Deutscher Ausschuß für Stahlbeton. Beuth,
Berlin, 1991.
[35] S.-H. Chen. Hydraulic Structures. Springer, Berlin, Heidelberg, 2015.
[36] S. Z. Cheng, S. J. Xu, R. Cong, and K. L. Wang. Experimental Study on

Mix Design of Plastic Concret with Lower Elastic Modulus and Higher Im-
pervious Coefficient. Advanced Materials Research, 535-537:1936–1939,
2012.
[37] Comité Européen de Normalisation. EN 12350-10:2010-12 - Prüfung von

Frischbeton - Teil 10: Selbstverdichtender Beton - L-Kasten-Versuch (Ger-
man version).

Frischbeton - Teil 2: Setzmaß (German version).

Frischbeton - Teil 5: Ausbreitmaß (German version).

Frischbeton - Teil 6: Frischbetonrohdichte (German version).

Festbeton - Teil 13: Bestimmung des Elastizitätsmoduls unter Druckbelas-
tung (Sekantenmodul) (German version).
61

Festbeton - Teil 2: Herstellung und Lagerung von Probekörpern für Fes-
tigkeitsprüfungen (German version).
[43] Comité Européen de Normalisation. EN 12390-2:2015-12/A20 - Prüfung

von Festbeton - Teil 2: Herstellung und Lagerung von Probekörpern für
Festigkeitsprüfungen; Änderung A20 (German version).

Festbeton - Teil 3: Druckfestigkeit von Probekörpern (German version).

Festbeton - Teil 4: Bestimmung der Druckfestigkeit; Anforderungen an Prüf-
maschinen (German version).

Festbeton - Teil 8: Wassereindringtiefe unter Druck (German version).
[47] Comité Européen de Normalisation. EN 1538:2015-10 - Ausführung von

Arbeiten im Spezialtiefbau – Schlitzwände (German version).
[48] Comité Européen de Normalisation. EN 197-1:2011-11 - Zement – Teil

1: Zusammensetzung, Anforderungen und Konformitätskriterien von Nor-
malzement (German version).
[49] Comité Européen de Normalisation. EN 1992-1-1:2011-01 - Bemessung

und Konstruktion von Stahlbeton- und Spannbetontragwerken - Teil 1-1:
Allgemeine Bemessungsregeln und Regeln für den Hochbau (German ver-
sion).
[50] Comité Européen de Normalisation. EN 206:2017-01 - Beton – Festlegung,

Eigenschaften, Herstellung und Konformität (German version).
[51] Comité Européen de Normalisation. EN ISO 10426-2:2006-02 - Erdöl- und

Erdgasindustrie - Zemente und Materialien für die Zementation von Tief-
bohrungen - Teil 2: Prüfung von Bohrloch-Zemente (German version).
[52] Comité Européen de Normalisation. EN ISO 10769:2011-09 - Geosyn-

thetische Tondichtungsbahnen – Bestimmung der Wasseraufnahme von
Bentonit (German version).
[53] J. Dahms and K. Rendchen. Über das Abgleichen der Druckflächen von
Würfeln und Zylindern aus Beton, Teil 1. Beton, 25(5):170–174, 1975.
[54] J. Dahms and K. Rendchen. Über das Abgleichen der Druckflächen von
Würfeln und Zylindern aus Beton, Teil 2. Beton, 25(6):209–212, 1975.
[55] J. de Vries. Afdichtingswanden van cement-bentoniet, rapport BSW 91-13.

Bouwdienst Rijkswaterstaat, Utrecht, 1992.
[56] Deutsche Vereinigung für Wasserwirtschaft, Abwasser und Abfall. Merk-

blatt 512-1 - Dichtungssysteme im Wasserbau - Teil 1: Erdbauwerke,
2012/02.
62
[57] Deutsches Institut für Normung. DIN 18122-1:1997-07 - Baugrund, Unter-

suchung von Bodenproben - Zustandsgrenzen (Konsistenzgrenzen). Teil 1:
Bestimmung der Fließ- und Ausrollgrenze.

suchung von Bodenproben - Bestimmung des Wasserdurchlässigkeitsbei-
werts. Teil 1: Laborversuche.

suchung von Bodenproben - Bestimmung des Wasserdurchlässigkeitsbei-
werts. Teil 2: Feldversuche.
[60] Deutsches Institut für Normung. DIN 18132:2012-04 - Baugrund, Versuche

und Versuchsgeräte – Bestimmung des Wasseraufnahmevermögens.
[61] Deutsches Institut für Normung. DIN 18135:2012-04 - Baugrund, Unter-

suchung von Bodenproben - Eindimensionaler Kompressionsversuch.
[62] Deutsches Institut für Normung. DIN 18136:2003-11 - Baugrund, Unter-

suchung von Bodenproben - Einaxialer Druckversuch.
[63] Deutsches Institut für Normung. DIN 18137-1:2010-07 Baugrund, Unter-

suchung von Bodenproben - Bestimmung der Scherfestigkeit. Teil 1: Be-
griffe und grundsätzliche Versuchsbedingungen.
[64] Deutsches Institut für Normung. DIN 18137-2:2011-04 Baugrund, Unter-

suchung von Bodenproben - Bestimmung der Scherfestigkeit. Teil 2: Triax-
ialversuch.
[65] Deutsches Institut für Normung. DIN 4093:2015-11 - Bemessung von ver-
festigten Bodenkörpern - Hergestellt mit Düsenstrahl-, Deep-Mixing- oder
Injektions-Verfahren.
[66] Deutsches Institut für Normung. DIN 4127:2014-02 - Erd- und Grundbau –
Prüfverfahren für Stützflüssigkeiten im Schlitzwandbau und für deren Aus-
gangsstoffe.
[67] J. Dietrich and C. Hotz. Bindemittelbasierte Bodenstabilisierung in der

Geotechnik. geotechnik, 40(4):271–278, 2017.
[68] C. Doby, E. Vuillermet, and M. Laugier. La Ganguise - Une paroi en beton

plastique pour etancher les rives d’un barrage. Travaux, (819):50–52, 2005.
[69] C. K. Edvardsen. Wasserdurchlässigkeit und Selbstheilung von Trennris-

sen in Beton, volume 455 of Deutscher Ausschuß für Stahlbeton. Beuth,
Berlin, 1996.
[70] European Federation of Foundation Contractors and Deep Foundations In-

stitute. Best Practice Guide to Tremie Concrete for Deep Foundations, 1st
Edition, 2016.
[71] European Federation of Foundation Contractors and Deep Foundations In-

stitute. Guide to Tremie Concrete for Deep Foundations, 2nd Edition, 2018.
63
[72] J. C. Evans, E. D. Stahl, and E. Drooff. Plastic Concrete Cutoff Walls. In

R. D. Woods (Ed.), Geotechnical practice for waste disposal ’87, Geotech-
nical Special Publications, pp. 462–472. ASCE, New York, 1987.
[73] R. Fell, P. MacGregor, D. Stapledon, and G. Bell. Geotechnical engineering

of dams. Taylor & Francis/Balkema, Boca Raton, FL, 1st Edition, 2005.
[74] R. Fell, P. MacGregor, D. Stapledon, G. Bell, and M. Foster. Geotechnical

engineering of dams. CRC Press/Balkema, Boca Raton, FL, 2nd Edition,
2015.
[75] J. W. France, P. A. Dickinson, S. J. Rigbey, I. A. Alvi, J. Trojanowski, and

H. T. Falvey. Interim Status Memorandum on Oroville Dam Spillway Inci-
dent, 05.09.2017.
[76] M. Geil. Untersuchungen der physikalischen und chemischen Eigen-

schaften von Bentonit-Zement-Suspensionen im frischen und erhärteten
Zustand. Dissertation, Technische Universität Braunschweig, Braun-
schweig, 1989.
[77] M. Ghazavi, Z. Safarzadeh, and H. Hashemolhoseini. Response Of Plas-

tic Concrete Cut-Off Walls In Earth Dams To Seismic Loading Using Finite
Element Methods. In Canadian Association for Earthquake Engineering
(Ed.), Proceedings of the 13th World Conference on Earthquake Engineer-
ing, Vancouver, 2004.
[78] R. E. Grim and N. Güven. Bentonites: Geology, mineralogy, properties and

uses, volume 24 of Developments in sedimentology. Elsevier, Amsterdam,
1978.
[79] P. Grübl, H. Weigler, and S. Karl. Beton: Arten, Herstellung und Eigen-
schaften. Handbuch fur Beton-, Stahlbeton- und Spannbetonbau. Ernst &
Sohn, Berlin, 2nd Edition, 2001.
[80] S. Hinchberger, J. Weck, and T. Newson. Mechanical and hydraulic charac-

terization of plastic concrete for seepage cut-off walls. Canadian Geotech-
nical Journal, 47(4):461–471, 2010.
[81] J. Hoffmann (Ed.). Handbuch der Messtechnik. Hanser, München, 4th

Edition, 2012.
[82] C. Högl. Bemessung von Tonbeton: Ermittlung des Einflusses von Baustof-
fkennwerten auf Basis eines geschlossenen Bemessungsmodells. Diplo-
marbeit, Technische Universität Wien, 2013.
[83] M. Hoseini, V. Bindiganavile, and N. Banthia. The effect of mechanical

stress on permeability of concrete: A review. Cement and Concrete Com-
posites, 31(4):213–220, 2009.
[84] S. Hoshino, T. Yamaguchi, T. Maeda, M. Mukai, T. Tanaka, and

S. Nakayama. Mineralogical Changes of Cement and Bentonite Accom-
panied With Their Interactions. In B. E. Burakov and A. S. Aloy (Eds.),
64
Proceedings of the 2009 MRS Meeting "Scientific Basis for Nuclear Waste
Management XXXIII", pp. 445–452, Warrendale, PA, 2009. Materials Re-
search Society.
[85] IBECO Bentonit-Technologie GmbH. Bentonit im Tiefbau: Ein Handbuch

für die Baupraxis. Mannheim, 1st Edition, 1998.
[86] Institution of Civil Engineers, Construction Industry Research and Informa-

tion Association, and Building Research Establishment. Specification for
the construction of slurry trench cut-off walls: as barriers to pollution migra-
tion, 1999.
[87] International Commission On Large Dams. Filling materials for watertight

cut off walls, Bulletin 51, 1985.
[88] International Federation for Structural Concrete (Ed.). fib Model Code for
Concrete Structures 2010. Ernst & Sohn, Berlin, 1st Edition, 2013.
[89] F. Jafarzadeh and S. H. Mousavi. Effect of Specimen’s Age on Mechan-

ical Properties of Plastic Concrete Walls in Dam Foundations. Electronic
Journal of Geotechnical Engineering, 17(D):473–482, 2012.
[90] K. Jasmund and G. Lagaly (Eds.). Tonminerale und Tone: Struktur, Eigen-
schaften, Anwendungen und Einsatz in Industrie und Umwelt. Steinkopff,
Darmstadt, 1993.
[91] T. W. Kahl, J. L. Kauschinger, and E. B. Perry. Plastic Concrete Cut-Off

Walls for Earth Dams, Technical Report REMR-GT-15. USACE, Vicksburg,
MS, 1991.
[92] S. Kazemian and S. Ghareh. Effects of Cement, Different Bentonite, and

Aggregates on Plastic Concrete in Besh-Ghardash Dam, Iran. Journal of
Testing and Evaluation, 45(1):242–248, 2017.
[93] S. Kazemian, S. Ghareh, and L. Torkanloo. To Investigation of Plastic Con-

crete Bentonite Changes on it’s Physical Properties. Procedia Engineering,
145:1080–1087, 2016.
[94] B. L. Kilpatrick and S. J. Garner. Use of cement-bentonite for cutoff wall

construction. In R. H. Borden, R. D. Holtz, and I. Juran (Eds.), Proceedings
of the 1992 American Society of Civil Engineers; Specialty Conference on
Grouting, Soil Improvement and Geosynthetics, Geotechnical Special Pub-
lication No. 30, pp. 803–815, ASME, New York, 1992.
[95] S. Klima and K. Beckhaus. Dichtwand im Sylvensteindamm. In S. A. Sa-

vidis and F. Remspecher (Eds.), Vorträge zum 10. Hans Lorenz Sympo-
sium, Veröffentlichungen des Grundbauinstitutes der Technischen Univer-
sität Berlin, pp. 235–243. Shaker, Aachen, 2014.
[96] D. Koch. Bentonitvergütete Abdichtungen. In K. Eichler (Ed.), Spezialtief-

bau, Kontakt & Studium, pp. 208–236. Expert Verlag, Renningen, 2015.
65
[97] D. Koch. Eigenschaften und Einsatzmöglichkeiten von Bentonit/Zement-

Mischungen im Speziealtiefbau. In K. Eichler (Ed.), Spezialtiefbau, Kontakt
& Studium, pp. 170–189. Expert Verlag, Renningen, 2015.
[98] T. Kränkel, D. Lowke, and C. Gehlen. Rheology Testing of Deep Foundation

Concrete. In W. Kusterle, O. Teubert, and M. Greim (Eds.), Rheologische
Messungen an Baustoffen 2016, pp. 211–223. tredition, Hamburg, 2016.
[99] P. Lappenküper. Untersuchungen zur Wasserdurchlässigkeit von Erdbe-

tondichtwänden unter Biegescherbeanspruchung. Masterarbeit, Technis-
che Universität Hamburg-Harburg, 2016.
[100] S.-H. Liu, L.-J. Wang, Z.-J. Wang, and E. Bauer. Numerical stress-
deformation analysis of cut-off wall in clay-core rockfill dam on thick over-
burden. Water Science and Engineering, 9(3):219–226, 2016.
[101] A. Mahboubi and A. Ajorloo. Experimental study of the mechanical behavior

of plastic concrete in triaxial compression. Cement and Concrete Research,
35(2):412–419, 2005.
[102] V. Malárics. Ermittlung der Betonzugfestigkeit aus dem Spaltzugversuch

an zylindrischen Betonproben: Dissertation, volume 69 of Karlsruher Reihe
Massivbau, Baustofftechnologie, Materialprüfung. KIT Scientific Publish-
ing, Karlsruhe, 2011.
[103] E. B. Nelson (Ed.). Well Cementing. Schlumberger, Sugar Land, Texas,

2nd Edition, 2006.
[104] R. Nevárez-Garibaldi, D. Miller, and G. Filz. Influence of Test Conditions

and Mixture Proportions on the Strength and Stiffness of Soil Treated with
Cement to Represent the Wet Method of Deep Mixing: A report to the Deep
Foundations Institute: In Progress. Virginia Tech, Blacksburg, 2017.
[105] A. M. Neville. Properties of concrete. Pearson, Harlow, 5th Edition, 2011.
[106] K.-H. Ohrdorf and H. Flachberger. Alkalische Aktivierung von Ben-

toniten als Beitrag der Aufbereitung zur Ressourcen- und Rohstoffeffizienz.
Chemie Ingenieur Technik, 82(11):1991–1998, 2010.
[107] K.-H. Ohrdorf, S. Kaufhold, F. Rüßmann, K. Ufer, and H. Flachberger. The

unusual behavior of a Milos bentonite in cement suspensions. Applied Clay
Science, 54(3-4):189–195, 2011.
[108] Österreichisches Normungsinstitut. ÖNORM B 4452 - Erd- und Grundbau

- Dichtwände im Untergrund, 1998/12.
[109] C. Ozyildirim and N. J. Carino. Concrete Strength Testing. In J. F. Lamond

and J. H. Pielert (Eds.), Significance of Tests and Properties of Concrete
and Concrete-Making Materials - STP 169D, pp. 125–140. ASTM Interna-
tional, West Conshohocken, PA, 2006.
66
[110] Y. Pashang Pisheh and S. M. Mir Mohammad Hosseini. Stress-strain be-

havior of plastic concrete using monotonic triaxial compression tests. Jour-
nal of Central South University, 19(4):1125–1131, 2012.
[111] J. Plank. Einfluss von Tonmineralien auf die Wirkung von PCE-Fließmitteln.
Beton, 66(7+8):288–290, 2016.
[112] S. Praetorius and B. Schößer. Bentonithandbuch: Ringspaltschmierung für

den Rohrvortrieb. Bauingenieur-Praxis. Ernst & Sohn, Berlin, 2016.
[113] G. M. Reeves, I. Sims, and J. C. Cripps (Eds.). Clay Materials Used in

Construction, volume 21 of Engineering Geology Special Publication. The
Geological Society, London, 2006.
[114] H.-W. Reinhardt. Ingenieurbaustoffe. Ernst & Sohn, Berlin, 2nd Edition,
2010.
[115] N. Roussel (Ed.). Understanding the rheology of concrete. Woodhead

Publishing, Cambridge and Philadelphia, PA, 2012.
[116] J. Sadrekarimi. Plastic Concrete Mechanical Behaviour. Journal of the In-

stitution of Engineers. India. Civil Engineering Division, 82(FEV):201–207,
2002.
[117] H.-H. Schmidt, R. F. Buchmaier, and C. Vogt-Breyer. Grundlagen der

Geotechnik: Geotechnik nach Eurocode. Springer, Wiesbaden, 5th Edi-
tion, 2017.
[118] C. Scholz, M. Rosenberg, J. Dietrich, and A. Märten. Durchlässigkeitsbes-

timmung von Einphasen-Dichtwandmassen - was ist theoretisch möglich
und was ist praktisch notwendig? In M. Rosenberg (Ed.), Qual-
itätssicherung und Innovation, Mitteilung des Instituts für Grundbau und
Bodenmechanik, Technische Universität Braunschweig, vol. 69, pp. 241–
263. Braunschweig, 2002.
[119] S. Seng and H. Tanaka. Properties of cement-treated soils during initial

curing stages. Soils and Foundations, 51(5):775–784, 2011.
[120] D. Sfer, I. Carol, R. Gettu, and G. Etse. Study of the Behavior of Con-
crete under Triaxial Compression. Journal of Engineering Mechanics,
128(2):156–163, 2002.
[121] K. Speck and M. Curbach. Ein einheitliches dreiaxiales Bruchkriterium für

alle Betone. Beton- und Stahlbetonbau, 105(4):233–243, 2010.
[122] Standardization Administration of the People’s Republic of China. DL/T

5150-2001 - Test code for hydraulic concrete, 2001.
[123] K. Terzaghi, R. B. Peck, and G. Mesri. Soil mechanics in engineering prac-

tice. Wiley, New Delhi, 3rd Edition, 1996.
[124] T. Triantafyllidis. Schlitzwand- und Dichtwandtechnik, volume 1 of Planung

und Bauausführung im Spezialtiefbau. Ernst & Sohn, Berlin, 2004.
67
[125] U.S. Bureau of Reclamation. Design Standards No. 13: Embankment Dam,
Chapter 16: Cutoff Walls, Revision 14, 2014/07.
[126] USACE. National Inventory of Dams (NID): Database - Corps Map:

http://nid.usace.army.mil/, 2017/12.
[127] R. D. Verástegui Flores, G. Di Emidio, and W. F. van Impe. Small-Strain

Shear Modulus and Strength Increase of Cement-Treated Clay. Geotech-
nical Testing Journal, 33(1):62–71, 2010.
[128] P. von Soos and J. Engel. Eigenschaften von Boden und Fels – ihre Ermit-
tlung im Labor. In K. J. Witt (Ed.), Grundbau-Taschenbuch, Teil 1: Geotech-
nische Grundlagen, pp. 123–218. Ernst & Sohn, Berlin, 2008.
[129] R. Weber, P. Bilgeri, H. Kollo, and H.-W. Vißmann (Eds.). Hochofenzement

- Eigenschaften und Anwendungen im Beton. Verl. Bau+Technik, Düssel-
dorf, 2nd Edition, 1998.
[130] A. Weinmann. Vertical Barriers with Plastic Concrete for Containment of

Landfills and Contaminated Land. In 3rd International Symposium and Ex-
hibition on Environmental Contamination in Central and Eastern Europe,
Warsaw, 1996.
[131] K. J. Witt (Ed.). Grundbau-Taschenbuch, Teil 1: Geotechnische Grundla-

gen. Ernst & Sohn, Berlin, 7th Edition, 2008.
[132] K. J. Witt (Ed.). Grundbau-Taschenbuch, Teil 2: Geotechnische Verfahren.

Ernst & Sohn, Berlin, 7th Edition, 2009.
[133] P. P. Xanthakos. Slurry walls. McGraw-Hill, New York, 1979.
[134] Y. Yang, H. Tong, S. Xu, and Y. Han. Effects of load level on water per-
meability of concrete. In W. Sun, editor, Microstructure related durability
of cementitious composites, RILEM proceedings PRO 61, pp. 545–552.
RILEM Publications, Bagneux, 2008.
[135] P. Zhang, Q. Guan, and Q. Li. Mechanical Properties of Plastic Concrete

Containing Bentonite. Research Journal of Applied Sciences, Engineering
and Technology, 5(4):1317–1322, 2013.
[136] H. Zwisler. Tonbeton, ein Baustoff für die Zukunft. Wasserwirtschaft,

74(11):542–544, 1984.
68

STAR - Plastic Concrete For Cut-Off Walls PDF

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

STAR - Plastic Concrete For Cut-Off Walls PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

STAR - Plastic Concrete For Cut-Off Walls PDF

Uploaded by

Copyright:

Available Formats

State-of-the-Art Report

Plastic Concrete for Cut-Off Walls

Karlsruhe Institute of Technology (KIT)

David Alós-Shepherd, M.Sc.

KIT – The Research University in the Helmholtz Association

Alós Shepherd, D.; Kotan, E.; Dehn, F. (2018):

5 Summary and Outlook 53

Figure 1.1: Schematic of an Earth Dam with Cut-Off Wall

amongst others, time, cost, geographical location, technological and geological

1.2 Definition & Field of Application

cement-bentonite backfilling [72]. In addition, the high strain capacity of Plastic

1.3 Cut-Off Wall Construction

The excavating method performance and choice greatly depends on depth,

This working slurry is then cleaned (de-sanded) or replaced by a fresh slurry

Compressive Strength Requirements

ICOLD Bulletin No. 51 recommends the elastic modulus of Plastic Concrete to be

Hydraulic Conductivity Requirements

Regarding the hydraulic conductivity of Plastic Concrete, most standards and

1.5 Problem Defintion

Contemporary standard concrete is considered a five-phase construction material

Some studies on the structure of single-phase diaphragm wall materials (such

(a) Ordinary Portland cement (b) Blast-furnace cement

Figure 2.1: SEM images of the micro-structure of hardened cement-bentonite mixtures

Structure & Properties

Figure 2.2: Schematic illustration of montmorillonite structure [112]

over, water adsorption does not occur instantaneously. Ca-bentonite adsorbs

a determining parameter for cement-stability [107]. It is therefore imperative to

solid semi-solid plastic liquid

Figure 2.3: Overview of soil states and corresponding water content

where IA : activity index (-)

Furthermore within the context of specialist foundation construction it may be

2.1.3 Aggregates and Admixtures

of tremie concrete [130]. Moreover, the type of aggregate used is regulated by

2.2 Mixture Composition

erally speaking, the lower bentonite contents correspond to Na-bentonite, whilst

Figure 2.4: Representative examples of Plastic Concrete mix designs

As can be seen, the Plastic Concrete mix design is a combination of standard

2.3 Mixing Sequence

Figure 2.5: Representative examples of Plastic Concrete mixing sequences

2.4 Fresh Properties

German DIN-Standard English ASTM-Standard

Especially important is the precise description of the mechanical behaviour of

3.1 Testing Standards

3.1.1 Compressive Strength

water pressure u = 0, as shown in Equation 3.1. In drained experiments water

where σ0 : effective stress (MPa)

3.1.2 Elastic Modulus

where EC,0 : initial elastic modulus (MPa)

It should be noted however, that the Oedometer modulus is dependant on the

Table 3.1: DIN and ASTM Test Method Standards Comparison

Field Test DIN-Standard ASTM-Standard

All in all it should be recommended to use concrete testing standards, since

therefore be specified during planning and tendering of projects to avoid testing

3.1.3 Testing Influences

Specimen end conditions

It is also questionable however, whether specimen capping is actually neces-

It is common knowledge that concrete is a crack afflicted material [114]. Therefore

be subjected to high scatter or even impossible. Measurement accuracy should